Edit photos by describing the change, on your own GPU. A free, open-source alternative to Nano Banana and ChatGPT image editing: photos stay on your computer, no daily limit. One installer for Windows, Mac and Linux sets up the best open model (Qwen-Image 2.1). MCP server for AI agents.
TypeScript ai-photo-editorcomfyui-alternativegenerative-fillimage-editingimage-generationinpaintinglinuxlocal-aimacosmcpmcp-servernano-banana-alternativeofflinephoto-editorprivacyqwen-imageqwen-image-2-1stable-diffusion-cppvulkanwindows
Site for 0.12: hero wall, video carousel, Layers lab demo, /image-to-layers/
Gabriele Lozuponea3cbc6a2f0d2
diff --git a/site/ai-image-editing-from-your-phone/index.html b/site/ai-image-editing-from-your-phone/index.htmlindex 773b888..f072a1f 100644--- a/site/ai-image-editing-from-your-phone/index.html+++ b/site/ai-image-editing-from-your-phone/index.html@@ -22,7 +22,7 @@<link rel="preconnect" href="https://fonts.googleapis.com"><link rel="preconnect" href="https://fonts.gstatic.com" crossorigin><link href="https://fonts.googleapis.com/css2?family=Inter:opsz,[email protected],400..700&family=JetBrains+Mono:wght@400;500&display=swap" rel="stylesheet">-<link rel="stylesheet" href="../assets/style.css?v=13">+<link rel="stylesheet" href="../assets/style.css?v=14"><script type="application/ld+json">{"@context": "https://schema.org",@@ -169,7 +169,8 @@<div class="dd-panel" id="dd-product"><ul><li><a href="/#examples"><b>Examples</b><span>Real edits, before and after</span></a></li><li><a href="/#features"><b>Features</b><span>Edit, paint, combine, expand, upscale</span></a></li>- <li><a href="/#devices"><b>On your phone <span class="new">New</span></b><span>Use your GPU from any device</span></a></li>+ <li><a href="/#layers"><b>Layers lab <span class="new">New</span></b><span>Split an image, move the pieces</span></a></li>+ <li><a href="/#devices"><b>On your phone</b><span>Use your GPU from any device</span></a></li><li><a href="/mcp-image-generation/"><b>For AI agents</b><span>MCP server and HTTP API</span></a></li><li><a href="/#requirements"><b>Requirements</b><span>GPUs, Macs and disk space</span></a></li></ul></div></li>@@ -179,6 +180,7 @@<li><a href="/self-hosted/"><b>Phone and server</b><span>Other devices, Linux, Docker, Tailscale</span></a></li><li><a href="/prompt-guide/"><b>Prompt guide</b><span>Write prompts that work</span></a></li><li><a href="/remove-objects-from-photos-offline/"><b>Remove objects</b><span>Step by step, offline</span></a></li>+ <li><a href="/image-to-layers/"><b>Image to layers <span class="new">New</span></b><span>Split a photo, export a PSD</span></a></li></ul></div></li><li class="dd"><button type="button" aria-expanded="false" aria-controls="dd-compare">Compare<svg class="caret" viewBox="0 0 12 12" aria-hidden="true"><path d="M3 4.5 6 7.5 9 4.5"/></svg></button><div class="dd-panel" id="dd-compare"><ul>@@ -292,6 +294,7 @@<a href="../comfyui-alternative/">Simple ComfyUI alternative</a><a href="../ai-image-editing-from-your-phone/">AI photo editing from your phone</a><a href="../remove-objects-from-photos-offline/">Remove objects offline</a>+ <a href="../image-to-layers/">Image to layers</a><a href="../qwen-image-2-1-local/">Run Qwen-Image locally</a><a href="../press/">Press kit</a><a href="../support/">Support Kokon</a>diff --git a/site/assets/img/layers/background.webp b/site/assets/img/layers/background.webpnew file mode 100644index 0000000..b8b42e9Binary files /dev/null and b/site/assets/img/layers/background.webp differdiff --git a/site/assets/img/layers/bats.webp b/site/assets/img/layers/bats.webpnew file mode 100644index 0000000..bc3163cBinary files /dev/null and b/site/assets/img/layers/bats.webp differdiff --git a/site/assets/img/layers/ghosts.webp b/site/assets/img/layers/ghosts.webpnew file mode 100644index 0000000..aff1dc6Binary files /dev/null and b/site/assets/img/layers/ghosts.webp differdiff --git a/site/assets/img/layers/girl.webp b/site/assets/img/layers/girl.webpnew file mode 100644index 0000000..f1c72ffBinary files /dev/null and b/site/assets/img/layers/girl.webp differdiff --git a/site/assets/img/layers/photo.webp b/site/assets/img/layers/photo.webpnew file mode 100644index 0000000..6f633d4Binary files /dev/null and b/site/assets/img/layers/photo.webp differdiff --git a/site/assets/img/layers/spider.webp b/site/assets/img/layers/spider.webpnew file mode 100644index 0000000..8371f51Binary files /dev/null and b/site/assets/img/layers/spider.webp differdiff --git a/site/assets/img/layers/text.webp b/site/assets/img/layers/text.webpnew file mode 100644index 0000000..47a01bbBinary files /dev/null and b/site/assets/img/layers/text.webp differdiff --git a/site/assets/img/videos/demo.webp b/site/assets/img/videos/demo.webpnew file mode 100644index 0000000..218df7eBinary files /dev/null and b/site/assets/img/videos/demo.webp differdiff --git a/site/assets/img/videos/update-0.12.webp b/site/assets/img/videos/update-0.12.webpnew file mode 100644index 0000000..2b818d7Binary files /dev/null and b/site/assets/img/videos/update-0.12.webp differdiff --git a/site/assets/img/videos/usecases.webp b/site/assets/img/videos/usecases.webpnew file mode 100644index 0000000..7183357Binary files /dev/null and b/site/assets/img/videos/usecases.webp differdiff --git a/site/assets/img/wall/bakery-text.webp b/site/assets/img/wall/bakery-text.webpnew file mode 100644index 0000000..27026cfBinary files /dev/null and b/site/assets/img/wall/bakery-text.webp differdiff --git a/site/assets/img/wall/bakery.webp b/site/assets/img/wall/bakery.webpnew file mode 100644index 0000000..aae26a1Binary files /dev/null and b/site/assets/img/wall/bakery.webp differdiff --git a/site/assets/img/wall/cabin-winter.webp b/site/assets/img/wall/cabin-winter.webpnew file mode 100644index 0000000..15b37abBinary files /dev/null and b/site/assets/img/wall/cabin-winter.webp differdiff --git a/site/assets/img/wall/cabin.webp b/site/assets/img/wall/cabin.webpnew file mode 100644index 0000000..b13ab43Binary files /dev/null and b/site/assets/img/wall/cabin.webp differdiff --git a/site/assets/img/wall/lighthouse-framed.webp b/site/assets/img/wall/lighthouse-framed.webpnew file mode 100644index 0000000..4a9acbcBinary files /dev/null and b/site/assets/img/wall/lighthouse-framed.webp differdiff --git a/site/assets/img/wall/lighthouse-wide.webp b/site/assets/img/wall/lighthouse-wide.webpnew file mode 100644index 0000000..7ac1fccBinary files /dev/null and b/site/assets/img/wall/lighthouse-wide.webp differdiff --git a/site/assets/img/wall/potter-watercolor.webp b/site/assets/img/wall/potter-watercolor.webpnew file mode 100644index 0000000..1a3b438Binary files /dev/null and b/site/assets/img/wall/potter-watercolor.webp differdiff --git a/site/assets/img/wall/potter.webp b/site/assets/img/wall/potter.webpnew file mode 100644index 0000000..fbaa124Binary files /dev/null and b/site/assets/img/wall/potter.webp differdiff --git a/site/assets/img/wall/room-sofa.webp b/site/assets/img/wall/room-sofa.webpnew file mode 100644index 0000000..df629c9Binary files /dev/null and b/site/assets/img/wall/room-sofa.webp differdiff --git a/site/assets/img/wall/room.webp b/site/assets/img/wall/room.webpnew file mode 100644index 0000000..05d3f2cBinary files /dev/null and b/site/assets/img/wall/room.webp differdiff --git a/site/assets/img/wall/sneaker-ad.webp b/site/assets/img/wall/sneaker-ad.webpnew file mode 100644index 0000000..794d1eaBinary files /dev/null and b/site/assets/img/wall/sneaker-ad.webp differdiff --git a/site/assets/img/wall/sneaker.webp b/site/assets/img/wall/sneaker.webpnew file mode 100644index 0000000..5d04ceeBinary files /dev/null and b/site/assets/img/wall/sneaker.webp differdiff --git a/site/assets/img/wall/street-cat.webp b/site/assets/img/wall/street-cat.webpnew file mode 100644index 0000000..6940d95Binary files /dev/null and b/site/assets/img/wall/street-cat.webp differdiff --git a/site/assets/img/wall/street-removed.webp b/site/assets/img/wall/street-removed.webpnew file mode 100644index 0000000..bf60706Binary files /dev/null and b/site/assets/img/wall/street-removed.webp differdiff --git a/site/assets/img/wall/street.webp b/site/assets/img/wall/street.webpnew file mode 100644index 0000000..b8d4addBinary files /dev/null and b/site/assets/img/wall/street.webp differdiff --git a/site/assets/img/wall/uc01-after.webp b/site/assets/img/wall/uc01-after.webpnew file mode 100644index 0000000..fd049b1Binary files /dev/null and b/site/assets/img/wall/uc01-after.webp differdiff --git a/site/assets/img/wall/uc01-before.webp b/site/assets/img/wall/uc01-before.webpnew file mode 100644index 0000000..5d1464dBinary files /dev/null and b/site/assets/img/wall/uc01-before.webp differdiff --git a/site/assets/img/wall/uc03-after.webp b/site/assets/img/wall/uc03-after.webpnew file mode 100644index 0000000..a05a964Binary files /dev/null and b/site/assets/img/wall/uc03-after.webp differdiff --git a/site/assets/img/wall/uc03-before.webp b/site/assets/img/wall/uc03-before.webpnew file mode 100644index 0000000..431e1feBinary files /dev/null and b/site/assets/img/wall/uc03-before.webp differdiff --git a/site/assets/img/wall/uc03-color.webp b/site/assets/img/wall/uc03-color.webpnew file mode 100644index 0000000..c01fb91Binary files /dev/null and b/site/assets/img/wall/uc03-color.webp differdiff --git a/site/assets/img/wall/uc04-after.webp b/site/assets/img/wall/uc04-after.webpnew file mode 100644index 0000000..ede5df7Binary files /dev/null and b/site/assets/img/wall/uc04-after.webp differdiff --git a/site/assets/img/wall/uc04-before.webp b/site/assets/img/wall/uc04-before.webpnew file mode 100644index 0000000..f9284aaBinary files /dev/null and b/site/assets/img/wall/uc04-before.webp differdiff --git a/site/assets/img/wall/uc05-after.webp b/site/assets/img/wall/uc05-after.webpnew file mode 100644index 0000000..851b5c7Binary files /dev/null and b/site/assets/img/wall/uc05-after.webp differdiff --git a/site/assets/img/wall/uc05-before.webp b/site/assets/img/wall/uc05-before.webpnew file mode 100644index 0000000..2c09e1eBinary files /dev/null and b/site/assets/img/wall/uc05-before.webp differdiff --git a/site/assets/img/wall/uc10-after.webp b/site/assets/img/wall/uc10-after.webpnew file mode 100644index 0000000..fd3c920Binary files /dev/null and b/site/assets/img/wall/uc10-after.webp differdiff --git a/site/assets/img/wall/uc10-before.webp b/site/assets/img/wall/uc10-before.webpnew file mode 100644index 0000000..a18feeaBinary files /dev/null and b/site/assets/img/wall/uc10-before.webp differdiff --git a/site/assets/img/wall/uc11-after.webp b/site/assets/img/wall/uc11-after.webpnew file mode 100644index 0000000..f38ce13Binary files /dev/null and b/site/assets/img/wall/uc11-after.webp differdiff --git a/site/assets/img/wall/uc11-before.webp b/site/assets/img/wall/uc11-before.webpnew file mode 100644index 0000000..a18feeaBinary files /dev/null and b/site/assets/img/wall/uc11-before.webp differdiff --git a/site/assets/img/wall/uc12-after.webp b/site/assets/img/wall/uc12-after.webpnew file mode 100644index 0000000..0cff023Binary files /dev/null and b/site/assets/img/wall/uc12-after.webp differdiff --git a/site/assets/img/wall/uc12-before.webp b/site/assets/img/wall/uc12-before.webpnew file mode 100644index 0000000..daefe6aBinary files /dev/null and b/site/assets/img/wall/uc12-before.webp differdiff --git a/site/assets/img/wall/uc13-after.webp b/site/assets/img/wall/uc13-after.webpnew file mode 100644index 0000000..389ce2dBinary files /dev/null and b/site/assets/img/wall/uc13-after.webp differdiff --git a/site/assets/img/wall/uc13-before.webp b/site/assets/img/wall/uc13-before.webpnew file mode 100644index 0000000..ddc857fBinary files /dev/null and b/site/assets/img/wall/uc13-before.webp differdiff --git a/site/assets/img/wall/uc16-after.webp b/site/assets/img/wall/uc16-after.webpnew file mode 100644index 0000000..02ca917Binary files /dev/null and b/site/assets/img/wall/uc16-after.webp differdiff --git a/site/assets/img/wall/uc16-before.webp b/site/assets/img/wall/uc16-before.webpnew file mode 100644index 0000000..7b08a1dBinary files /dev/null and b/site/assets/img/wall/uc16-before.webp differdiff --git a/site/assets/img/wall/uc17-after.webp b/site/assets/img/wall/uc17-after.webpnew file mode 100644index 0000000..e78be41Binary files /dev/null and b/site/assets/img/wall/uc17-after.webp differdiff --git a/site/assets/img/wall/uc17-before.webp b/site/assets/img/wall/uc17-before.webpnew file mode 100644index 0000000..c481e75Binary files /dev/null and b/site/assets/img/wall/uc17-before.webp differdiff --git a/site/assets/layers-demo.js b/site/assets/layers-demo.jsnew file mode 100644index 0000000..3646cbb--- /dev/null+++ b/site/assets/layers-demo.js@@ -0,0 +1,197 @@+// The Layers lab demo (kokon.app home + /image-to-layers/): the Halloween poster from the 0.12 video+// and the real layers Kokon made from it (cut-outs + the background it painted behind them). Split+// opens the stack in 3D, then each object can be dragged, hidden or shown. Positions are percent of+// the 1328 px square poster; the spider layer is wider than the frame (it was completed with Expand).+(() => {+ const root = document.querySelector("[data-layers-demo]");+ if (!root) return;+ const SRC = "/assets/img/layers/";+ // Bottom to top, as in the app's layer list.+ const LAYERS = [+ { id: "text", name: "Text", x: 19.8, y: 2.11, w: 59.94, h: 20.18 },+ { id: "bats", name: "Bats", x: 15.29, y: 4.14, w: 69.65, h: 32.98 },+ { id: "ghosts", name: "Ghosts", x: 8.36, y: 38.1, w: 79.67, h: 20.56 },+ { id: "spider", name: "Spider", x: 62.88, y: 75.83, w: 46.99, h: 24.17 },+ { id: "girl", name: "Girl", x: 28.09, y: 23.49, w: 41.64, h: 69.73 },+ ];+ const reduced = matchMedia("(prefers-reduced-motion: reduce)").matches;+ const eye = (on) =>+ on+ ? '<svg viewBox="0 0 24 24"><path d="M2 12s3.6-7 10-7 10 7 10 7-3.6 7-10 7S2 12 2 12z"/><circle cx="12" cy="12" r="3"/></svg>'+ : '<svg viewBox="0 0 24 24"><path d="M3 3l18 18M10.6 5.1A9.6 9.6 0 0 1 12 5c6.4 0 10 7 10 7a17 17 0 0 1-3 3.8M6.6 6.6C3.7 8.4 2 12 2 12s3.6 7 10 7a9.5 9.5 0 0 0 5.4-1.6"/><path d="M9.9 9.9a3 3 0 0 0 4.2 4.2"/></svg>';++ root.innerHTML = `+ <div class="ld-stage">+ <div class="ld-scene clip" aria-label="A Halloween poster that can be split into layers">+ <img class="ld-layer bg" src="${SRC}background.webp" alt="" style="--z:0" draggable="false">+ ${LAYERS.map((l, i) => `<img class="ld-layer" data-id="${l.id}" src="${SRC}${l.id}.webp" alt="" draggable="false" style="--z:${i + 1};left:${l.x}%;top:${l.y}%;width:${l.w}%;height:${l.h}%">`).join("")}+ <img class="ld-photo" src="${SRC}photo.webp" alt="A Halloween poster: a girl with a pumpkin, ghosts, bats, a spider and the words Happy Halloween" draggable="false">+ </div>+ <div class="ld-flash"></div>+ <div class="ld-hint" role="status"></div>+ <button class="btn primary small ld-go" type="button">Split into layers</button>+ </div>+ <div class="ld-side">+ <div class="ld-panel">+ <h3>Layers <small>Experimental</small></h3>+ <div class="ld-rows"><div class="ld-empty">One image. Press Split into layers.</div></div>+ </div>+ <div class="ld-export">+ <h3>Export to PSD</h3>+ <p>Every layer in a PSD for Photoshop, GIMP or Photopea.</p>+ <button class="btn ghost small ld-psd" type="button" disabled>Export PSD</button>+ </div>+ </div>`;+ const scene = root.querySelector(".ld-scene");+ const go = root.querySelector(".ld-go");+ const hint = root.querySelector(".ld-hint");+ const rows = root.querySelector(".ld-rows");+ const flash = root.querySelector(".ld-flash");+ const psd = root.querySelector(".ld-psd");+ psd.addEventListener("click", () => say("In Kokon this saves a PSD with these six layers"));+ const el = Object.fromEntries(LAYERS.map((l) => [l.id, scene.querySelector(`[data-id="${l.id}"]`)]));+ const state = Object.fromEntries(LAYERS.map((l) => [l.id, { x: l.x, y: l.y, hidden: false }]));+ let selected = null;+ let split = false;++ // Alpha masks (small) to pick the object under the pointer, not its bounding box.+ const masks = {};+ LAYERS.forEach((l) => {+ const im = new Image();+ im.src = SRC + l.id + ".webp";+ im.decode().then(() => {+ const c = document.createElement("canvas");+ c.width = 160;+ c.height = Math.max(1, Math.round((160 * im.naturalHeight) / im.naturalWidth));+ const g = c.getContext("2d", { willReadFrequently: true });+ g.drawImage(im, 0, 0, c.width, c.height);+ masks[l.id] = { w: c.width, h: c.height, data: g.getImageData(0, 0, c.width, c.height).data };+ }, () => {});+ });+ const hit = (px, py) => {+ for (let i = LAYERS.length - 1; i >= 0; i--) {+ const l = LAYERS[i], s = state[l.id], m = masks[l.id];+ if (s.hidden) continue;+ const u = (px - s.x) / l.w, v = (py - s.y) / l.h;+ if (u < 0 || v < 0 || u >= 1 || v >= 1) continue;+ if (!m) return l.id;+ if (m.data[(Math.floor(v * m.h) * m.w + Math.floor(u * m.w)) * 4 + 3] > 40) return l.id;+ }+ return null;+ };++ const place = (id) => { el[id].style.left = state[id].x + "%"; el[id].style.top = state[id].y + "%"; el[id].classList.toggle("hidden", state[id].hidden); };+ const select = (id) => {+ selected = id;+ LAYERS.forEach((l) => el[l.id].classList.toggle("sel", l.id === id));+ rows.querySelectorAll(".ld-row").forEach((r) => r.classList.toggle("sel", r.dataset.id === id));+ };+ const say = (text) => { hint.textContent = text; hint.classList.toggle("on", !!text); };++ function drawRows() {+ rows.innerHTML =+ [...LAYERS].reverse().map((l) => `<div class="ld-row" data-id="${l.id}"><img src="${SRC}${l.id}.webp" alt=""><span>${l.name}</span><button class="ld-eye" type="button" aria-label="Hide ${l.name}">${eye(true)}</button></div>`).join("") ++ `<div class="ld-row"><img src="${SRC}background.webp" alt=""><span>Background</span></div><button class="btn ghost small ld-reset" type="button">Back to the photo</button>`;+ rows.querySelectorAll(".ld-row[data-id]").forEach((r) => {+ const id = r.dataset.id;+ r.addEventListener("click", () => select(id));+ r.querySelector(".ld-eye").addEventListener("click", (e) => {+ e.stopPropagation();+ state[id].hidden = !state[id].hidden;+ place(id);+ e.currentTarget.innerHTML = eye(!state[id].hidden);+ e.currentTarget.setAttribute("aria-label", `${state[id].hidden ? "Show" : "Hide"} ${LAYERS.find((l) => l.id === id).name}`);+ });+ });+ rows.querySelector(".ld-reset").addEventListener("click", reset);+ }++ const wait = (ms) => new Promise((r) => setTimeout(r, reduced ? 0 : ms));+ async function doSplit() {+ if (split) return;+ split = true;+ go.remove();+ scene.classList.add("split");+ drawRows();+ if (!reduced) {+ say("Each object on its own layer, the background painted in behind");+ // The photo fades out while the picture is still flat; only then does it open in 3D.+ await wait(550);+ scene.classList.remove("clip");+ await new Promise((r) => requestAnimationFrame(() => requestAnimationFrame(r)));+ scene.classList.add("open");+ await wait(2400);+ say("");+ scene.classList.add("landing");+ scene.classList.remove("open");+ await wait(1900);+ scene.classList.remove("landing");+ scene.classList.add("clip");+ flash.classList.remove("on");+ void flash.offsetWidth;+ flash.classList.add("on");+ await wait(500);+ }+ psd.disabled = false;+ scene.classList.add("ready");+ // Show once what a move does: the spider slides in from the edge, whole (completed with Expand).+ say("Drag anything: the street behind it is already there");+ select("spider");+ const from = state.spider.x, to = 26, t0 = performance.now(), dur = reduced ? 1 : 1400;+ await new Promise((done) => {+ const step = (now) => {+ const t = Math.min(1, (now - t0) / dur), e = t < 0.5 ? 4 * t * t * t : 1 - Math.pow(-2 * t + 2, 3) / 2;+ state.spider.x = from + (to - from) * e;+ place("spider");+ t < 1 ? requestAnimationFrame(step) : done();+ };+ requestAnimationFrame(step);+ });+ await wait(2600);+ say("");+ }+ async function reset() {+ scene.classList.remove("ready");+ select(null);+ say("");+ LAYERS.forEach((l) => { el[l.id].style.transition = "left 0.7s cubic-bezier(0.65, 0, 0.35, 1), top 0.7s cubic-bezier(0.65, 0, 0.35, 1), opacity 0.4s"; state[l.id] = { x: l.x, y: l.y, hidden: false }; place(l.id); });+ await wait(750);+ LAYERS.forEach((l) => (el[l.id].style.transition = ""));+ split = false;+ psd.disabled = true;+ scene.classList.remove("split", "open");+ rows.innerHTML = '<div class="ld-empty">One image. Press Split into layers.</div>';+ root.querySelector(".ld-stage").appendChild(go);+ }+ go.addEventListener("click", doSplit);++ // Drag: pointer position in percent of the poster.+ let drag = null;+ const pct = (e) => { const r = scene.getBoundingClientRect(); return [((e.clientX - r.left) / r.width) * 100, ((e.clientY - r.top) / r.height) * 100]; };+ scene.addEventListener("pointerdown", (e) => {+ if (!scene.classList.contains("ready") || e.target.closest("button")) return;+ const [px, py] = pct(e);+ const id = hit(px, py);+ select(id);+ if (!id) return;+ say("");+ drag = { id, px, py, x: state[id].x, y: state[id].y };+ scene.classList.add("dragging");+ scene.setPointerCapture(e.pointerId);+ });+ scene.addEventListener("pointermove", (e) => {+ if (!drag) return;+ const [px, py] = pct(e), l = LAYERS.find((q) => q.id === drag.id);+ // Keep at least a fifth of the object inside the frame.+ state[drag.id].x = Math.max(-l.w * 0.8, Math.min(100 - l.w * 0.2, drag.x + px - drag.px));+ state[drag.id].y = Math.max(-l.h * 0.8, Math.min(100 - l.h * 0.2, drag.y + py - drag.py));+ place(drag.id);+ });+ const end = () => { drag = null; scene.classList.remove("dragging"); };+ scene.addEventListener("pointerup", end);+ scene.addEventListener("pointercancel", end);++ // Split by itself the first time it comes into view; after that it is the visitor's.+ const io = new IntersectionObserver(([en]) => { if (en.isIntersecting) { io.disconnect(); setTimeout(doSplit, 900); } }, { threshold: 0.6 });+ io.observe(scene);+})();diff --git a/site/assets/style.css b/site/assets/style.cssindex 7328017..902767f 100644--- a/site/assets/style.css+++ b/site/assets/style.css@@ -96,9 +96,9 @@ html.menu-lock, html.menu-lock body { overflow: hidden; }.btn.primary:hover { transform: translateY(-1px); box-shadow: 0 0 0 1px rgba(255,255,255,0.3), 0 14px 50px rgba(247,246,242,0.2); }.btn.ghost { background: var(--card-hi); box-shadow: inset 0 0 0 1px var(--line-hi); }.btn.ghost:hover { background: rgba(255,255,255,0.1); }-.btn.yt { background: #ff0033; color: #fff; box-shadow: 0 10px 40px rgba(255,0,51,0.25); }-.btn.yt svg { fill: currentColor; }-.btn.yt:hover { transform: translateY(-1px); background: #e6002e; box-shadow: 0 14px 50px rgba(255,0,51,0.35); }+.btn.yt { background: var(--card-hi); box-shadow: inset 0 0 0 1px var(--line-hi); }+.btn.yt svg { fill: #e5484d; }+.btn.yt:hover { background: rgba(255,255,255,0.1); }.btn.small { height: 38px; padding: 0 16px; font-size: 14px; }.btn .tag { font-size: 11.5px; font-weight: 600; padding: 2px 7px; border-radius: 6px; background: rgba(255,255,255,0.1); color: var(--text-2); letter-spacing: 0.02em; }@@ -393,3 +393,109 @@ footer a:hover { color: var(--text); }.scene.phones { width: 604px; height: 573px; }.article figure .scene-box img { border-radius: 0; box-shadow: none; }.sec.devices-top { padding-top: 96px; }++/* ---------- Hero (0.12): the first screen says what Kokon is; behind it, before/after pairs made by Kokon,+ tilted and scrolling as in the launch film's opening. The videos follow right below. ---------- */+.hero { padding-top: 0; }+.hero-top { min-height: 100svh; display: flex; flex-direction: column; align-items: center; justify-content: center; padding: 96px 0 56px; }+/* Above the page grain (body::before, z-index 100), like the video frame: in Chrome a translucent layer over+ composited content (the 3D wall promotes it) greys it. The nav (102) stays on top. */+.hero .wrap { position: relative; z-index: 101; }+.hero-top h1 { text-shadow: 0 4px 40px rgba(0,0,0,0.7); }+.hero-top .fine { color: var(--text-2); text-shadow: 0 2px 12px rgba(0,0,0,0.9); }+.hero-top .ctas .btn.ghost { background: rgba(16,16,16,0.86); backdrop-filter: blur(10px); -webkit-backdrop-filter: blur(10px); box-shadow: inset 0 0 0 1px rgba(255,255,255,0.2); }+.hero-top .ctas .btn.ghost:hover { background: rgba(32,32,32,0.92); }+.hero-top .eyebrow { background: rgba(12,12,12,0.82); backdrop-filter: blur(10px); -webkit-backdrop-filter: blur(10px); }+.wall { position: absolute; z-index: 0; left: 0; right: 0; top: 0; height: calc(100svh + 140px); overflow: hidden; pointer-events: none; perspective: 1800px; -webkit-mask-image: linear-gradient(#000 70%, transparent); mask-image: linear-gradient(#000 70%, transparent); }+/* Dark only behind the words, so the pictures keep their colour around them. */+.wall::after { content: ""; position: absolute; inset: 0; background: radial-gradient(ellipse 46% 34% at 50% 46%, rgba(7,7,7,0.86) 0%, rgba(7,7,7,0.6) 55%, rgba(7,7,7,0.04) 100%); }+.wall-plane { position: absolute; left: 50%; top: 40%; display: flex; flex-direction: column; gap: 22px; transform: translate(-50%, -50%) rotateX(22deg) rotateZ(-9deg); }+.wall-row { display: flex; gap: 22px; width: max-content; animation: wall-l 120s linear infinite; will-change: transform; }+.wall-row.r { animation-name: wall-r; }+@keyframes wall-l { from { transform: translateX(0); } to { transform: translateX(-50%); } }+@keyframes wall-r { from { transform: translateX(-50%); } to { transform: translateX(0); } }+.wall-tile { position: relative; flex: none; width: 360px; aspect-ratio: 3 / 2; border-radius: 16px; overflow: hidden; background: #111; box-shadow: 0 20px 60px rgba(0,0,0,0.5); }+.wall-tile img { position: absolute; inset: 0; width: 100%; height: 100%; object-fit: cover; }+/* The wipe from before to after: the after's box slides in while its picture slides back (transforms only). */+.wall-tile .aw { position: absolute; inset: 0; overflow: hidden; transform: translateX(100%); animation: wipe-box 10s cubic-bezier(0.65, 0, 0.35, 1) infinite; animation-delay: var(--d); }+.wall-tile .aw img { transform: translateX(-100%); animation: wipe-img 10s cubic-bezier(0.65, 0, 0.35, 1) infinite; animation-delay: var(--d); }+.wall-tile .aw::before { content: ""; position: absolute; left: 0; top: 0; bottom: 0; width: 2px; z-index: 1; background: #fff; box-shadow: 0 0 12px rgba(255,255,255,0.8); }+@keyframes wipe-box { 0%, 20% { transform: translateX(100%); } 32%, 82% { transform: translateX(0); } 94%, 100% { transform: translateX(100%); } }+@keyframes wipe-img { 0%, 20% { transform: translateX(-100%); } 32%, 82% { transform: translateX(0); } 94%, 100% { transform: translateX(-100%); } }+.wall-tile .chip { position: absolute; left: 10px; bottom: 10px; z-index: 2; max-width: calc(100% - 20px); padding: 5px 11px; border-radius: 999px; background: rgba(0,0,0,0.62); font-size: 13px; color: #f2f1ed; white-space: nowrap; overflow: hidden; text-overflow: ellipsis; box-shadow: inset 0 0 0 1px rgba(255,255,255,0.14); }+.hero .showreel { margin-top: 0; }+@media (max-width: 719px) { .wall-tile { width: 230px; } .wall-tile .chip { font-size: 11px; } .wall-plane, .wall-row { gap: 14px; } .wall::after { background: radial-gradient(ellipse 70% 34% at 50% 46%, rgba(7,7,7,0.86) 0%, rgba(7,7,7,0.6) 55%, rgba(7,7,7,0.2) 100%); } }+@media (prefers-reduced-motion: reduce) { .wall-row, .wall-tile .aw, .wall-tile .aw img { animation: none; } .wall-tile:nth-child(even) .aw, .wall-tile:nth-child(even) .aw img { transform: none; } }++/* ---------- Video carousel: the launch film (local file), then YouTube videos loaded only on click ---------- */+.showreel .slide { position: absolute; inset: 0; opacity: 0; visibility: hidden; transition: opacity 0.45s, visibility 0s 0.45s; }+.showreel .slide.on { opacity: 1; visibility: visible; transition: opacity 0.45s; }+.showreel .slide iframe { width: 100%; height: 100%; border: 0; display: block; }+.yt-facade { position: absolute; inset: 0; width: 100%; height: 100%; background: #000 center / cover no-repeat; display: grid; place-items: center; }+.yt-facade .play { width: 84px; height: 84px; border-radius: 50%; background: rgba(0,0,0,0.62); backdrop-filter: blur(8px); box-shadow: inset 0 0 0 1px rgba(255,255,255,0.22); display: grid; place-items: center; transition: transform 0.2s, background 0.2s; }+.yt-facade .play::before { content: ""; margin-left: 6px; border-left: 26px solid #fff; border-top: 16px solid transparent; border-bottom: 16px solid transparent; }+.yt-facade:hover .play { transform: scale(1.06); background: rgba(229,72,77,0.85); }+.yt-facade .yt-note { position: absolute; left: 18px; bottom: 16px; font-size: 13px; color: rgba(255,255,255,0.78); background: rgba(0,0,0,0.55); padding: 6px 12px; border-radius: 999px; }+/* Above the video frame (z-index 101): its big dark shadow fell over the tabs and greyed them. */+.vtabs { position: relative; z-index: 102; display: grid; grid-template-columns: repeat(3, 1fr); gap: 12px; margin-top: 16px; }+/* The video shown has a light border and brighter text; a line fills along its bottom while it plays. */+.vtab { position: relative; overflow: hidden; display: flex; align-items: center; gap: 12px; padding: 8px; border-radius: 16px; background: var(--card); box-shadow: inset 0 0 0 1px var(--line); text-align: left; transition: background 0.2s, box-shadow 0.2s; }+.vtab:hover { background: var(--card-hi); }+.vtab.on { background: var(--card-hi); box-shadow: inset 0 0 0 1.5px rgba(247,246,242,0.7); }+.vtab.on b { color: #fff; }+.vtab .bar { position: absolute; left: 0; bottom: 0; height: 2px; width: 0; background: var(--paper); opacity: 0; }+.vtab.on .bar { opacity: 0.8; }+.vtab img { width: 96px; aspect-ratio: 16 / 9; object-fit: cover; border-radius: 9px; flex: none; }+.vtab b { display: block; font-size: 14.5px; font-weight: 600; letter-spacing: -0.01em; }+.vtab span { display: block; font-size: 12.5px; color: var(--text-3); margin-top: 2px; }+@media (max-width: 719px) { .vtabs { grid-template-columns: 1fr; gap: 8px; } .vtab img { width: 76px; } }++/* ---------- 0.12: the Layers lab demo (the Halloween poster from the 0.12 video, its real layers) ---------- */+.layers-demo { display: grid; grid-template-columns: minmax(0, 1fr) 280px; gap: 22px; max-width: 900px; margin: 56px auto 0; text-align: left; }+.ld-stage { position: relative; perspective: 1600px; }+.ld-scene { position: relative; aspect-ratio: 1; border-radius: 18px; transform-style: preserve-3d; transition: transform 1.2s cubic-bezier(0.16, 1, 0.3, 1); box-shadow: 0 0 0 1px var(--line-hi), 0 40px 120px rgba(0,0,0,0.6); touch-action: none; user-select: none; -webkit-user-select: none; }+/* Clipped only while flat: clip-path flattens the 3D stack, so it comes back when the layers have landed. */+.ld-scene.clip { clip-path: inset(0 round 18px); }+.ld-scene.open { box-shadow: none; transform: rotateX(54deg) rotateZ(-32deg) scale(0.74); }+/* Each layer always 1 px above the one below: coplanar layers flicker in 3D (z-fighting), and no easing overshoots below 0. */+.ld-layer { position: absolute; pointer-events: none; transform: translateZ(calc(var(--z) * 1px)); transition: transform 1.1s cubic-bezier(0.22, 1, 0.36, 1), opacity 0.4s, filter 0.2s; }+.ld-layer.bg, .ld-photo { left: 0; top: 0; width: 100%; height: 100%; border-radius: 18px; }+/* Lifting: the top layers first. Landing: one after the other from the bottom. */+.ld-scene.open .ld-layer { transform: translateZ(calc(var(--z) * 56px)); transition-delay: calc((5 - var(--z)) * 60ms); transition-timing-function: cubic-bezier(0.16, 1, 0.3, 1); }+.ld-scene.landing .ld-layer { transition-delay: calc(var(--z) * 120ms); }+.ld-scene.open .ld-layer.bg { box-shadow: 0 0 0 1px rgba(255,255,255,0.25); }+.ld-photo { position: absolute; transition: opacity 0.5s; z-index: 10; }+.ld-scene.split .ld-photo { opacity: 0; pointer-events: none; }+.ld-layer.hidden { opacity: 0; }+.ld-layer.sel { filter: drop-shadow(0 0 1.5px #fff) drop-shadow(0 0 1.5px #fff) drop-shadow(0 6px 18px rgba(0,0,0,0.5)); }+.ld-scene.split.ready { cursor: grab; }+.ld-scene.dragging { cursor: grabbing; }+/* A soft light passes over the picture when the layers have landed. */+.ld-flash { position: absolute; inset: 0; z-index: 15; border-radius: 18px; pointer-events: none; opacity: 0; background: linear-gradient(105deg, transparent 30%, rgba(255,255,255,0.28) 50%, transparent 70%) -100% 0 / 200% 100% no-repeat; }+.ld-flash.on { animation: ld-flash 0.9s ease-out; }+@keyframes ld-flash { 0% { opacity: 1; background-position: 120% 0; } 100% { opacity: 1; background-position: -120% 0; } }+.ld-go { position: absolute; left: 50%; bottom: 22px; z-index: 20; transform: translateX(-50%); box-shadow: 0 10px 40px rgba(0,0,0,0.5); }+.ld-hint { position: absolute; left: 50%; bottom: 16px; z-index: 20; transform: translateX(-50%); padding: 7px 14px; border-radius: 999px; background: rgba(0,0,0,0.66); backdrop-filter: blur(8px); font-size: 13.5px; white-space: nowrap; opacity: 0; transition: opacity 0.4s; pointer-events: none; }+.ld-hint.on { opacity: 1; }+/* The right column: the layer list, then Export to PSD, together as tall as the picture. */+.ld-side { display: flex; flex-direction: column; gap: 14px; min-height: 0; }+.ld-panel, .ld-export { border-radius: 18px; background: var(--card); box-shadow: inset 0 0 0 1px var(--line); padding: 14px; }+.ld-panel { flex: 1; display: flex; flex-direction: column; gap: 4px; min-height: 0; overflow: auto; }+.ld-panel h3, .ld-export h3 { margin: 0 0 6px; font-size: 14px; font-weight: 600; display: flex; align-items: center; justify-content: space-between; }+.ld-panel h3 small { font-weight: 500; font-size: 11px; letter-spacing: 0.04em; text-transform: uppercase; color: #f2c14e; background: rgba(242,193,78,0.12); padding: 2px 8px; border-radius: 999px; }+.ld-rows { display: flex; flex-direction: column; gap: 4px; flex: 1; }+.ld-row { display: flex; align-items: center; gap: 10px; padding: 7px 8px; border-radius: 10px; font-size: 14px; color: var(--text-2); cursor: pointer; }+.ld-row.sel { background: var(--card-hi); color: var(--text); }+.ld-row img { width: 34px; height: 34px; object-fit: contain; border-radius: 6px; background: repeating-conic-gradient(#222 0 25%, #2c2c2c 0 50%) 0 0 / 10px 10px; flex: none; }+.ld-row span { flex: 1; }+.ld-eye { width: 30px; height: 30px; border-radius: 8px; display: grid; place-items: center; color: var(--text-3); }+.ld-eye:hover { background: rgba(255,255,255,0.08); color: var(--text); }+.ld-eye svg { width: 17px; height: 17px; fill: none; stroke: currentColor; stroke-width: 2; stroke-linecap: round; stroke-linejoin: round; }+.ld-panel .ld-empty { font-size: 13.5px; color: var(--text-3); padding: 6px 8px; }+.ld-panel .ld-reset { margin-top: auto; align-self: flex-start; }+.ld-export p { margin: 0 0 12px; font-size: 13.5px; color: var(--text-2); line-height: 1.45; }+.ld-export .btn { width: 100%; }+.ld-export .btn:disabled { opacity: 0.45; cursor: default; transform: none; }+.layers-ctas { display: flex; flex-wrap: wrap; gap: 12px; justify-content: center; margin-top: 34px; }+@media (max-width: 819px) { .layers-demo { grid-template-columns: 1fr; } .ld-panel { overflow: visible; } .ld-rows { flex-direction: row; flex-wrap: wrap; } .ld-row { padding: 4px 6px; } .ld-row img { width: 28px; height: 28px; } }+@media (prefers-reduced-motion: reduce) { .ld-scene, .ld-layer { transition: none; } .ld-flash.on { animation: none; } }diff --git a/site/assets/usecases.mp4 b/site/assets/usecases.mp4new file mode 100644index 0000000..fe3a35fBinary files /dev/null and b/site/assets/usecases.mp4 differdiff --git a/site/comfyui-alternative/index.html b/site/comfyui-alternative/index.htmlindex 1fde3bc..b5d6618 100644--- a/site/comfyui-alternative/index.html+++ b/site/comfyui-alternative/index.html@@ -22,7 +22,7 @@<link rel="preconnect" href="https://fonts.googleapis.com"><link rel="preconnect" href="https://fonts.gstatic.com" crossorigin><link href="https://fonts.googleapis.com/css2?family=Inter:opsz,[email protected],400..700&family=JetBrains+Mono:wght@400;500&display=swap" rel="stylesheet">-<link rel="stylesheet" href="../assets/style.css?v=13">+<link rel="stylesheet" href="../assets/style.css?v=14"><script type="application/ld+json">{"@context": "https://schema.org",@@ -126,7 +126,8 @@<div class="dd-panel" id="dd-product"><ul><li><a href="/#examples"><b>Examples</b><span>Real edits, before and after</span></a></li><li><a href="/#features"><b>Features</b><span>Edit, paint, combine, expand, upscale</span></a></li>- <li><a href="/#devices"><b>On your phone <span class="new">New</span></b><span>Use your GPU from any device</span></a></li>+ <li><a href="/#layers"><b>Layers lab <span class="new">New</span></b><span>Split an image, move the pieces</span></a></li>+ <li><a href="/#devices"><b>On your phone</b><span>Use your GPU from any device</span></a></li><li><a href="/mcp-image-generation/"><b>For AI agents</b><span>MCP server and HTTP API</span></a></li><li><a href="/#requirements"><b>Requirements</b><span>GPUs, Macs and disk space</span></a></li></ul></div></li>@@ -136,6 +137,7 @@<li><a href="/self-hosted/"><b>Phone and server</b><span>Other devices, Linux, Docker, Tailscale</span></a></li><li><a href="/prompt-guide/"><b>Prompt guide</b><span>Write prompts that work</span></a></li><li><a href="/remove-objects-from-photos-offline/"><b>Remove objects</b><span>Step by step, offline</span></a></li>+ <li><a href="/image-to-layers/"><b>Image to layers <span class="new">New</span></b><span>Split a photo, export a PSD</span></a></li></ul></div></li><li class="dd"><button type="button" aria-expanded="false" aria-controls="dd-compare">Compare<svg class="caret" viewBox="0 0 12 12" aria-hidden="true"><path d="M3 4.5 6 7.5 9 4.5"/></svg></button><div class="dd-panel" id="dd-compare"><ul>@@ -210,7 +212,7 @@</tbody></table></div>- <p>Speed with the default model: on an RTX 5070 Ti, about 24 seconds for an edit at 1.5K and about 36 seconds for a new image at native 2K. 12 GB cards are a little slower, 8 GB cards take 1 to 2 minutes per edit, and an M2 Pro Mac takes about 4 minutes per image.</p>+ <p>Speed with the default model: on an RTX 5070 Ti, about 24 seconds for an edit at 1.5K and about 36 seconds for a new image at native 2K. 12 GB cards are a little slower, 8 GB cards take 1 to 2 minutes per edit, and an M2 Pro Mac takes about a minute for a new 1-megapixel image (about 106 s for an edit) with Kokon's own Mac engine.</p><h2>Questions</h2><div class="faq">@@ -243,6 +245,7 @@<a href="../offline-alternative-to-nano-banana/">Offline Nano Banana alternative</a><a href="../comfyui-alternative/">Simple ComfyUI alternative</a><a href="../remove-objects-from-photos-offline/">Remove objects offline</a>+ <a href="../image-to-layers/">Image to layers</a><a href="../qwen-image-2-1-local/">Run Qwen-Image locally</a><a href="../press/">Press kit</a><a href="../support/">Support Kokon</a>diff --git a/site/huggingface/index.html b/site/huggingface/index.htmlindex fa0b3ad..88b85c7 100644--- a/site/huggingface/index.html+++ b/site/huggingface/index.html@@ -22,7 +22,7 @@<link rel="preconnect" href="https://fonts.googleapis.com"><link rel="preconnect" href="https://fonts.gstatic.com" crossorigin><link href="https://fonts.googleapis.com/css2?family=Inter:opsz,[email protected],400..700&family=JetBrains+Mono:wght@400;500&display=swap" rel="stylesheet">-<link rel="stylesheet" href="../assets/style.css?v=13">+<link rel="stylesheet" href="../assets/style.css?v=14"><style>.hf .hf-from { display: inline-block; margin: 0 0 14px; padding: 5px 12px; border: 1px solid var(--line-hi); border-radius: 12px; font-size: 13.5px; color: var(--text-3); }.hf .hf-from a { color: var(--text); }@@ -57,7 +57,8 @@<div class="dd-panel" id="dd-product"><ul><li><a href="/#examples"><b>Examples</b><span>Real edits, before and after</span></a></li><li><a href="/#features"><b>Features</b><span>Edit, paint, combine, expand, upscale</span></a></li>- <li><a href="/#devices"><b>On your phone <span class="new">New</span></b><span>Use your GPU from any device</span></a></li>+ <li><a href="/#layers"><b>Layers lab <span class="new">New</span></b><span>Split an image, move the pieces</span></a></li>+ <li><a href="/#devices"><b>On your phone</b><span>Use your GPU from any device</span></a></li><li><a href="/mcp-image-generation/"><b>For AI agents</b><span>MCP server and HTTP API</span></a></li><li><a href="/#requirements"><b>Requirements</b><span>GPUs, Macs and disk space</span></a></li></ul></div></li>@@ -67,6 +68,7 @@<li><a href="/self-hosted/"><b>Phone and server</b><span>Other devices, Linux, Docker, Tailscale</span></a></li><li><a href="/prompt-guide/"><b>Prompt guide</b><span>Write prompts that work</span></a></li><li><a href="/remove-objects-from-photos-offline/"><b>Remove objects</b><span>Step by step, offline</span></a></li>+ <li><a href="/image-to-layers/"><b>Image to layers <span class="new">New</span></b><span>Split a photo, export a PSD</span></a></li></ul></div></li><li class="dd"><button type="button" aria-expanded="false" aria-controls="dd-compare">Compare<svg class="caret" viewBox="0 0 12 12" aria-hidden="true"><path d="M3 4.5 6 7.5 9 4.5"/></svg></button><div class="dd-panel" id="dd-compare"><ul>diff --git a/site/image-to-layers/index.html b/site/image-to-layers/index.htmlnew file mode 100644index 0000000..b096478--- /dev/null+++ b/site/image-to-layers/index.html@@ -0,0 +1,331 @@+<!doctype html>+<html lang="en">+<head>+<meta charset="utf-8">+<meta name="viewport" content="width=device-width, initial-scale=1">+<title>Image to Layers: Split a Photo into PSD Layers, Free and Offline | Kokon</title>+<meta name="description" content="Click each object and Kokon cuts it onto its own layer and paints the background behind it. Move, hide or edit the layers, then export a layered PSD. Free, local and open source; your image is never uploaded.">+<meta name="author" content="Gabriele Lozupone">+<meta name="robots" content="index, follow, max-image-preview:large">+<meta name="theme-color" content="#0a0a0a">+<link rel="canonical" href="https://kokon.app/image-to-layers/">+<link rel="icon" href="../assets/favicon.svg" type="image/svg+xml">+<link rel="icon" href="../assets/favicon.ico" sizes="any">+<link rel="apple-touch-icon" href="../assets/apple-touch-icon.png">+<meta property="og:site_name" content="Kokon">+<meta property="og:type" content="article">+<meta property="og:title" content="Split an image into layers and export a PSD, offline and free">+<meta property="og:description" content="Click each object and Kokon cuts it onto its own layer and paints the background behind it. Move, hide or edit the layers, then export a layered PSD. Free, local and open source; your image is never uploaded.">+<meta property="og:url" content="https://kokon.app/image-to-layers/">+<meta property="og:image" content="https://kokon.app/assets/img/layers/photo.webp">+<meta name="twitter:card" content="summary_large_image">+<link rel="preconnect" href="https://fonts.googleapis.com">+<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>+<link href="https://fonts.googleapis.com/css2?family=Inter:opsz,[email protected],400..700&family=JetBrains+Mono:wght@400;500&display=swap" rel="stylesheet">+<link rel="stylesheet" href="../assets/style.css?v=14">+<script type="application/ld+json">+{+ "@context": "https://schema.org",+ "@graph": [+ {+ "@type": "TechArticle",+ "headline": "Image to Layers: Split a Photo into PSD Layers, Free and Offline | Kokon",+ "description": "Click each object and Kokon cuts it onto its own layer and paints the background behind it. Move, hide or edit the layers, then export a layered PSD. Free, local and open source; your image is never uploaded.",+ "image": "https://kokon.app/assets/img/layers/photo.webp",+ "datePublished": "2026-11-05",+ "dateModified": "2026-11-05",+ "author": {+ "@type": "Person",+ "name": "Gabriele Lozupone",+ "url": "https://gabrielelozupone.github.io/"+ },+ "about": [+ {+ "@type": "SoftwareApplication",+ "name": "Kokon",+ "url": "https://kokon.app/"+ }+ ],+ "mainEntityOfPage": "https://kokon.app/image-to-layers/",+ "video": {+ "@type": "VideoObject",+ "name": "Kokon 0.12: split any image into layers, move objects, export to PSD",+ "description": "The Layers lab in Kokon 0.12: objects cut out onto their own layers, the background filled, layers moved and edited, a PSD exported.",+ "thumbnailUrl": "https://kokon.app/assets/img/videos/update-0.12.webp",+ "uploadDate": "2026-11-04",+ "duration": "PT2M21S",+ "embedUrl": "https://www.youtube.com/embed/eGr9yYe7NVc"+ }+ },+ {+ "@type": "HowTo",+ "name": "How to split an image into layers and export a PSD with Kokon",+ "step": [+ {+ "@type": "HowToStep",+ "name": "Open the image",+ "text": "Drop the image into Kokon, then press Layers (lab) under the prompt box."+ },+ {+ "@type": "HowToStep",+ "name": "Click the objects",+ "text": "Click each thing you want on its own layer. Click again to take a bigger part, right-click to take a piece away, or fix the edge with the brush."+ },+ {+ "@type": "HowToStep",+ "name": "Split",+ "text": "Press Split into layers and choose Fast or Quality. Kokon cuts each object out and paints the background behind them."+ },+ {+ "@type": "HowToStep",+ "name": "Move and edit",+ "text": "Drag, scale, hide or edit each layer on its own; Expand a layer to complete an object cut off by the frame."+ },+ {+ "@type": "HowToStep",+ "name": "Export",+ "text": "Press Export PSD for a layered file, or Save as image for one picture."+ }+ ]+ },+ {+ "@type": "BreadcrumbList",+ "itemListElement": [+ {+ "@type": "ListItem",+ "position": 1,+ "name": "Kokon",+ "item": "https://kokon.app/"+ },+ {+ "@type": "ListItem",+ "position": 2,+ "name": "Image to layers",+ "item": "https://kokon.app/image-to-layers/"+ }+ ]+ },+ {+ "@type": "FAQPage",+ "mainEntity": [+ {+ "@type": "Question",+ "name": "Can AI split a flat JPG or PNG into separate layers?",+ "acceptedAnswer": {+ "@type": "Answer",+ "text": "Yes. In Kokon's Layers lab you click each thing you want on its own layer; Kokon cuts it out and paints the background behind it, so the result is a stack of layers: the objects on top, the filled background at the bottom."+ }+ },+ {+ "@type": "Question",+ "name": "How do I convert an image to a PSD with layers for free?",+ "acceptedAnswer": {+ "@type": "Answer",+ "text": "Open the image in Kokon, press Layers (lab), click the objects, press Split into layers, then Export PSD. Kokon is free and open source; the file is written on your computer."+ }+ },+ {+ "@type": "Question",+ "name": "What happens to the background behind an object I separate?",+ "acceptedAnswer": {+ "@type": "Answer",+ "text": "Kokon repaints it with the image model, continuing the surfaces, light and perspective around it, so moving or hiding the object leaves no hole."+ }+ },+ {+ "@type": "Question",+ "name": "Does it work offline, without uploading my image?",+ "acceptedAnswer": {+ "@type": "Answer",+ "text": "Yes. The cut-outs, the background fill and the PSD are all made on your own computer. After setup no internet is needed."+ }+ },+ {+ "@type": "Question",+ "name": "Can I edit the text in the image after the split?",+ "acceptedAnswer": {+ "@type": "Answer",+ "text": "Text in the picture becomes an image layer you can move, hide or edit with a sentence, not editable type."+ }+ },+ {+ "@type": "Question",+ "name": "Which devices does it run on?",+ "acceptedAnswer": {+ "@type": "Answer",+ "text": "Windows and Linux with a 12 GB graphics card (NVIDIA 8 GB in beta) and Apple Silicon Macs. The Layers lab is in the desktop app; it is off in the phone layout."+ }+ }+ ]+ }+ ]+}+</script>+</head>+<body>+<svg width="0" height="0" style="position:absolute" aria-hidden="true">+ <defs>+ <symbol id="mark" viewBox="0 0 1024 1024"><path d="M0.00 242.81 C0.00 112.17 70.23 25.08 175.59 25.08 C280.94 25.08 351.17 112.17 351.17 242.81 L351.17 781.19 C351.17 911.83 280.94 998.92 175.59 998.92 C70.23 998.92 0.00 911.83 0.00 781.19 Z M389.53 548.95 C389.53 250.35 662.35 25.08 1024.00 25.08 C1024.00 323.68 751.18 548.95 389.53 548.95 Z M389.53 587.31 C389.53 821.93 618.71 998.92 922.49 998.92 C922.49 764.30 693.31 587.31 389.53 587.31 Z" fill="currentColor"/></symbol>+ <symbol id="win" viewBox="0 0 24 24"><path d="M0 0h11.3v11.3H0zM12.7 0H24v11.3H12.7zM0 12.7h11.3V24H0zM12.7 12.7H24V24H12.7z"/></symbol>+ <symbol id="apple" viewBox="0 0 24 24"><path d="M12.152 6.896c-.948 0-2.415-1.078-3.96-1.04-2.04.027-3.91 1.183-4.961 3.014-2.117 3.675-.546 9.103 1.519 12.09 1.013 1.454 2.208 3.09 3.792 3.039 1.52-.065 2.09-.987 3.935-.987 1.831 0 2.35.987 3.96.948 1.637-.026 2.676-1.48 3.676-2.948 1.156-1.688 1.636-3.325 1.662-3.415-.039-.013-3.182-1.221-3.22-4.857-.026-3.04 2.48-4.494 2.597-4.559-1.429-2.09-3.623-2.324-4.39-2.376-2-.156-3.675 1.09-4.61 1.09zM15.53 3.83c.843-1.012 1.4-2.427 1.245-3.83-1.207.052-2.662.805-3.532 1.818-.78.896-1.454 2.338-1.273 3.714 1.338.104 2.715-.688 3.559-1.701"/></symbol>+ <symbol id="linux" viewBox="0 0 24 24"><path d="M12.504 0c-.155 0-.315.008-.48.021-4.226.333-3.105 4.807-3.17 6.298-.076 1.092-.3 1.953-1.05 3.02-.885 1.051-2.127 2.75-2.716 4.521-.278.832-.41 1.684-.287 2.489a.424.424 0 00-.11.135c-.26.268-.45.6-.663.839-.199.199-.485.267-.797.4-.313.136-.658.269-.864.68-.09.189-.136.394-.132.602 0 .199.027.4.055.536.058.399.116.728.04.97-.249.68-.28 1.145-.106 1.484.174.334.535.47.94.601.81.2 1.91.135 2.774.6.926.466 1.866.67 2.616.47.526-.116.97-.464 1.208-.946.587-.003 1.23-.269 2.26-.334.699-.058 1.574.267 2.577.2.025.134.063.198.114.333l.003.003c.391.778 1.113 1.132 1.884 1.071.771-.06 1.592-.536 2.257-1.306.631-.765 1.683-1.084 2.378-1.503.348-.199.629-.469.649-.853.023-.4-.2-.811-.714-1.376v-.097l-.003-.003c-.17-.2-.25-.535-.338-.926-.085-.401-.182-.786-.492-1.046h-.003c-.059-.054-.123-.067-.188-.135a.357.357 0 00-.19-.064c.431-1.278.264-2.55-.173-3.694-.533-1.41-1.465-2.638-2.175-3.483-.796-1.005-1.576-1.957-1.56-3.368.026-2.152.236-6.133-3.544-6.139zm.529 3.405h.013c.213 0 .396.062.584.198.19.135.33.332.438.533.105.259.158.459.166.724 0-.02.006-.04.006-.06v.105a.086.086 0 01-.004-.021l-.004-.024a1.807 1.807 0 01-.15.706.953.953 0 01-.213.335.71.71 0 00-.088-.042c-.104-.045-.198-.064-.284-.133a1.312 1.312 0 00-.22-.066c.05-.06.146-.133.183-.198.053-.128.082-.264.088-.402v-.02a1.21 1.21 0 00-.061-.4c-.045-.134-.101-.2-.183-.333-.084-.066-.167-.132-.267-.132h-.016c-.093 0-.176.03-.262.132a.8.8 0 00-.205.334 1.18 1.18 0 00-.09.4v.019c.002.089.008.179.02.267-.193-.067-.438-.135-.607-.202a1.635 1.635 0 01-.018-.2v-.02a1.772 1.772 0 01.15-.768c.082-.22.232-.406.43-.533a.985.985 0 01.594-.2zm-2.962.059h.036c.142 0 .27.048.399.135.146.129.264.288.344.465.09.199.14.4.153.667v.004c.007.134.006.2-.002.266v.08c-.03.007-.056.018-.083.024-.152.055-.274.135-.393.2.012-.09.013-.18.003-.267v-.015c-.012-.133-.04-.2-.082-.333a.613.613 0 00-.166-.267.248.248 0 00-.183-.064h-.021c-.071.006-.13.04-.186.132a.552.552 0 00-.12.27.944.944 0 00-.023.33v.015c.012.135.037.2.08.334.046.134.098.2.166.268.01.009.02.018.034.024-.07.057-.117.07-.176.136a.304.304 0 01-.131.068 2.62 2.62 0 01-.275-.402 1.772 1.772 0 01-.155-.667 1.759 1.759 0 01.08-.668 1.43 1.43 0 01.283-.535c.128-.133.26-.2.418-.2zm1.37 1.706c.332 0 .733.065 1.216.399.293.2.523.269 1.052.468h.003c.255.136.405.266.478.399v-.131a.571.571 0 01.016.47c-.123.31-.516.643-1.063.842v.002c-.268.135-.501.333-.775.465-.276.135-.588.292-1.012.267a1.139 1.139 0 01-.448-.067 3.566 3.566 0 01-.322-.198c-.195-.135-.363-.332-.612-.465v-.005h-.005c-.4-.246-.616-.512-.686-.71-.07-.268-.005-.47.193-.6.224-.135.38-.271.483-.336.104-.074.143-.102.176-.131h.002v-.003c.169-.202.436-.47.839-.601.139-.036.294-.065.466-.065zm2.8 2.142c.358 1.417 1.196 3.475 1.735 4.473.286.534.855 1.659 1.102 3.024.156-.005.33.018.513.064.646-1.671-.546-3.467-1.089-3.966-.22-.2-.232-.335-.123-.335.59.534 1.365 1.572 1.646 2.757.13.535.16 1.104.021 1.67.067.028.135.06.205.067 1.032.534 1.413.938 1.23 1.537v-.043c-.06-.003-.12 0-.18 0h-.016c.151-.467-.182-.825-1.065-1.224-.915-.4-1.646-.336-1.77.465-.008.043-.013.066-.018.135-.068.023-.139.053-.209.064-.43.268-.662.669-.793 1.187-.13.533-.17 1.156-.205 1.869v.003c-.02.334-.17.838-.319 1.35-1.5 1.072-3.58 1.538-5.348.334a2.645 2.645 0 00-.402-.533 1.45 1.45 0 00-.275-.333c.182 0 .338-.03.465-.067a.615.615 0 00.314-.334c.108-.267 0-.697-.345-1.163-.345-.467-.931-.995-1.788-1.521-.63-.4-.986-.87-1.15-1.396-.165-.534-.143-1.085-.015-1.645.245-1.07.873-2.11 1.274-2.763.107-.065.037.135-.408.974-.396.751-1.14 2.497-.122 3.854a8.123 8.123 0 01.647-2.876c.564-1.278 1.743-3.504 1.836-5.268.048.036.217.135.289.202.218.133.38.333.59.465.21.201.477.335.876.335.039.003.075.006.11.006.412 0 .73-.134.997-.268.29-.134.52-.334.74-.4h.005c.467-.135.835-.402 1.044-.7zm2.185 8.958c.037.6.343 1.245.882 1.377.588.134 1.434-.333 1.791-.765l.211-.01c.315-.007.577.01.847.268l.003.003c.208.199.305.53.391.876.085.4.154.78.409 1.066.486.527.645.906.636 1.14l.003-.007v.018l-.003-.012c-.015.262-.185.396-.498.595-.63.401-1.746.712-2.457 1.57-.618.737-1.37 1.14-2.036 1.191-.664.053-1.237-.2-1.574-.898l-.005-.003c-.21-.4-.12-1.025.056-1.69.176-.668.428-1.344.463-1.897.037-.714.076-1.335.195-1.814.12-.465.308-.797.641-.984l.045-.022zm-10.814.049h.01c.053 0 .105.005.157.014.376.055.706.333 1.023.752l.91 1.664.003.003c.243.533.754 1.064 1.189 1.637.434.598.77 1.131.729 1.57v.006c-.057.744-.48 1.148-1.125 1.294-.645.135-1.52.002-2.395-.464-.968-.536-2.118-.469-2.857-.602-.369-.066-.61-.2-.723-.4-.11-.2-.113-.602.123-1.23v-.004l.002-.003c.117-.334.03-.752-.027-1.118-.055-.401-.083-.71.043-.94.16-.334.396-.4.69-.533.294-.135.64-.202.915-.47h.002v-.002c.256-.268.445-.601.668-.838.19-.201.38-.336.663-.336zm7.159-9.074c-.435.201-.945.535-1.488.535-.542 0-.97-.267-1.28-.466-.154-.134-.28-.268-.373-.335-.164-.134-.144-.333-.074-.333.109.016.129.134.199.2.096.066.215.2.36.333.292.2.68.467 1.167.467.485 0 1.053-.267 1.398-.466.195-.135.445-.334.648-.467.156-.136.149-.267.279-.267.128.016.034.134-.147.332a8.097 8.097 0 01-.69.468zm-1.082-1.583V5.64c-.006-.02.013-.042.029-.05.074-.043.18-.027.26.004.063 0 .16.067.15.135-.006.049-.085.066-.135.066-.055 0-.092-.043-.141-.068-.052-.018-.146-.008-.163-.065zm-.551 0c-.02.058-.113.049-.166.066-.047.025-.086.068-.14.068-.05 0-.13-.02-.136-.068-.01-.066.088-.133.15-.133.08-.031.184-.047.259-.005.019.009.036.03.03.05v.02h.003z"/></symbol>+ <symbol id="youtube" viewBox="0 0 24 24"><path d="M23.5 6.19a3.02 3.02 0 0 0-2.12-2.14C19.5 3.55 12 3.55 12 3.55s-7.5 0-9.38.5A3.02 3.02 0 0 0 .5 6.19C0 8.07 0 12 0 12s0 3.93.5 5.81a3.02 3.02 0 0 0 2.12 2.14c1.88.5 9.38.5 9.38.5s7.5 0 9.38-.5a3.02 3.02 0 0 0 2.12-2.14C24 15.93 24 12 24 12s0-3.93-.5-5.81zM9.55 15.57V8.43L15.82 12l-6.27 3.57z"/></symbol>+ <symbol id="github" viewBox="0 0 24 24"><path d="M12 .297c-6.63 0-12 5.373-12 12 0 5.303 3.438 9.8 8.205 11.385.6.113.82-.258.82-.577 0-.285-.01-1.04-.015-2.04-3.338.724-4.042-1.61-4.042-1.61C4.422 18.07 3.633 17.7 3.633 17.7c-1.087-.744.084-.729.084-.729 1.205.084 1.838 1.236 1.838 1.236 1.07 1.835 2.809 1.305 3.495.998.108-.776.417-1.305.76-1.605-2.665-.3-5.466-1.332-5.466-5.93 0-1.31.465-2.38 1.235-3.22-.135-.303-.54-1.523.105-3.176 0 0 1.005-.322 3.3 1.23.96-.267 1.98-.399 3-.405 1.02.006 2.04.138 3 .405 2.28-1.552 3.285-1.23 3.285-1.23.645 1.653.24 2.873.12 3.176.765.84 1.23 1.91 1.23 3.22 0 4.61-2.805 5.625-5.475 5.92.42.36.81 1.096.81 2.22 0 1.606-.015 2.896-.015 3.286 0 .315.21.69.825.57C20.565 22.092 24 17.592 24 12.297c0-6.627-5.373-12-12-12"/></symbol>+ </defs>+</svg>++<header class="nav scrolled" id="nav">+ <div class="wrap">+ <a class="logo" href="../" aria-label="Kokon home"><svg><use href="#mark"/></svg>Kokon</a>+ <nav class="nav-links" id="nav-links" aria-label="Site">+ <ul>+ <li class="dd"><button type="button" aria-expanded="false" aria-controls="dd-product">Product<svg class="caret" viewBox="0 0 12 12" aria-hidden="true"><path d="M3 4.5 6 7.5 9 4.5"/></svg></button>+ <div class="dd-panel" id="dd-product"><ul>+ <li><a href="/#examples"><b>Examples</b><span>Real edits, before and after</span></a></li>+ <li><a href="/#features"><b>Features</b><span>Edit, paint, combine, expand, upscale</span></a></li>+ <li><a href="/#layers"><b>Layers lab <span class="new">New</span></b><span>Split an image, move the pieces</span></a></li>+ <li><a href="/#devices"><b>On your phone</b><span>Use your GPU from any device</span></a></li>+ <li><a href="/mcp-image-generation/"><b>For AI agents</b><span>MCP server and HTTP API</span></a></li>+ <li><a href="/#requirements"><b>Requirements</b><span>GPUs, Macs and disk space</span></a></li>+ </ul></div></li>+ <li class="dd"><button type="button" aria-expanded="false" aria-controls="dd-guides">Guides<svg class="caret" viewBox="0 0 12 12" aria-hidden="true"><path d="M3 4.5 6 7.5 9 4.5"/></svg></button>+ <div class="dd-panel" id="dd-guides"><ul>+ <li><a href="/qwen-image-2-1-local/"><b>Setup guide</b><span>Install and run Qwen-Image 2.1</span></a></li>+ <li><a href="/self-hosted/"><b>Phone and server</b><span>Other devices, Linux, Docker, Tailscale</span></a></li>+ <li><a href="/prompt-guide/"><b>Prompt guide</b><span>Write prompts that work</span></a></li>+ <li><a href="/remove-objects-from-photos-offline/"><b>Remove objects</b><span>Step by step, offline</span></a></li>+ <li><a href="/image-to-layers/"><b>Image to layers <span class="new">New</span></b><span>Split a photo, export a PSD</span></a></li>+ </ul></div></li>+ <li class="dd"><button type="button" aria-expanded="false" aria-controls="dd-compare">Compare<svg class="caret" viewBox="0 0 12 12" aria-hidden="true"><path d="M3 4.5 6 7.5 9 4.5"/></svg></button>+ <div class="dd-panel" id="dd-compare"><ul>+ <li><a href="/local-ai-photo-editor/"><b>Local AI photo editor</b><span>Why edit on your own computer</span></a></li>+ <li><a href="/offline-alternative-to-nano-banana/"><b>vs Nano Banana</b><span>Offline alternative to cloud editing</span></a></li>+ <li><a href="/comfyui-alternative/"><b>vs ComfyUI</b><span>Simpler, no nodes, no Python</span></a></li>+ <li><a href="/ai-image-editing-from-your-phone/"><b>From your phone</b><span>Kokon vs ComfyUI with --listen</span></a></li>+ </ul></div></li>+ <li><a class="top" href="/#faq">FAQ</a></li>+ </ul>+ <p class="nav-more"><a href="https://github.com/GabrieleLozupone/kokon" rel="noopener">GitHub</a><a href="/press/">Press kit</a></p>+ <a class="btn kofi-btn" href="https://ko-fi.com/gabrielelozupone" target="_blank" rel="noopener"><svg viewBox="0 0 24 24" aria-hidden="true"><path d="M12 20.5s-7.3-4.4-9.3-8.9C1.2 8.3 3.2 4.5 6.8 4.5c2.1 0 3.5 1.1 4.3 2.5h1.8c.8-1.4 2.2-2.5 4.3-2.5 3.6 0 5.6 3.8 4.1 7.1-2 4.5-9.3 8.9-9.3 8.9z"/></svg>Support Kokon on Ko-fi</a>+ </nav>+ <span class="spacer"></span>+ <a class="gh" href="https://github.com/GabrieleLozupone/kokon" rel="noopener"><svg width="18" height="18" fill="currentColor"><use href="#github"/></svg>GitHub</a>+ <a class="kofi" href="https://ko-fi.com/gabrielelozupone" target="_blank" rel="noopener"><svg viewBox="0 0 24 24" aria-hidden="true"><path d="M12 20.5s-7.3-4.4-9.3-8.9C1.2 8.3 3.2 4.5 6.8 4.5c2.1 0 3.5 1.1 4.3 2.5h1.8c.8-1.4 2.2-2.5 4.3-2.5 3.6 0 5.6 3.8 4.1 7.1-2 4.5-9.3 8.9-9.3 8.9z"/></svg>Support</a>+ <a class="btn primary small" href="https://github.com/GabrieleLozupone/kokon/releases/latest" rel="noopener" data-dl-auto>Download</a>+ <button class="burger" type="button" aria-expanded="false" aria-controls="nav-links" aria-label="Open menu"><span class="bars"><span></span><span></span></span><span class="lbl">Menu</span></button>+ </div>+</header>++<main class="article">+ <div class="wrap narrow">+ <nav class="crumbs" aria-label="Breadcrumb"><a href="../">Kokon</a> <span>/</span> Image to layers</nav>+ <h1>Split any image into layers, on your own computer</h1>+ <p class="lede-left">Click the things you want on their own layer. Kokon cuts each one out, paints the background behind it, and gives you a stack of layers to move, hide or edit, or to export as a PSD. Free, offline, and your image is never uploaded. New in Kokon 0.12, experimental. Updated November 5, 2026.</p>++ <p><strong>Try it here:</strong> this is a real split made by Kokon, the poster from the 0.12 video. Drag the girl, the spider, the ghosts or the text.</p>+ </div>+ <div class="wrap">+ <div class="layers-demo" data-layers-demo></div>+ </div>+ <div class="wrap narrow">+ <h2>How to split a photo into layers</h2>+ <ol>+ <li><strong>Open the image</strong> in Kokon (drop it into the window or paste it) and press <strong>Layers (lab)</strong> under the prompt box.</li>+ <li><strong>Click each thing</strong> you want on its own layer: a person, an object, the text. Click again to take a bigger part, right-click to take a piece away, or fix the edge with the brush (B) and the eraser (E).</li>+ <li>Press <strong>Split into layers</strong> and choose <strong>Fast</strong> (a quick cut-out, a few seconds per object) or <strong>Quality</strong> (the AI redraws each edge as a clean cut-out).</li>+ <li>Kokon <strong>paints the background</strong> behind the objects, so nothing leaves a hole when it moves.</li>+ <li><strong>Move, scale or hide</strong> any layer, select several with Shift, group them with Ctrl+G.</li>+ <li><strong>Export PSD</strong> for a layered file, or <strong>Save as image</strong> to get one picture back as a new version of your project.</li>+ </ol>++ <h2>The background behind each object gets filled</h2>+ <p>Cutting an object out is only half of it: what was behind it has to exist too. Kokon repaints the empty background with the same image model it uses for edits, continuing the surfaces, light and perspective around it. In the demo above, drag the spider away and the street is already there.</p>++ <h2>Export a layered PSD</h2>+ <p>Export PSD writes every layer, with its transparency and position, plus the flattened picture, in the Photoshop format that GIMP and Photopea also read. Keep going there: retouch one layer, add your own, change the order. Text in the picture comes out as an image layer, not as editable type.</p>++ <h2>Edit one layer, expand it, blend it in</h2>+ <ul>+ <li><strong>Edit one layer:</strong> select it and describe the change; only that layer changes.</li>+ <li><strong>Expand a layer:</strong> an object cut off by the frame can be extended and completed (the spider in the demo was).</li>+ <li><strong>Blend into scene</strong> (experimental): after moving things, Kokon reads where each object stands and draws its shadows, or reflections on wet or shiny ground.</li>+ <li><strong>Draw a new object</strong> with the pencil, one colour per object, and say what it is.</li>+ </ul>++ <h2>Local and free, or an online image-to-layers tool?</h2>+ <p>Online tools that turn an image into layers upload your picture, and most cap the free plan. Workflows for ComfyUI run locally but need nodes, Python and a large GPU. Kokon is one installer: the cut-outs, the fill and the PSD are made on your own graphics card, with no limit and no account. The images are for non-commercial use because of the model's license.</p>++ <div class="layers-ctas" style="justify-content:flex-start">+ <a class="btn yt" href="https://youtu.be/eGr9yYe7NVc" rel="noopener"><svg><use href="#youtube"/></svg>Watch it in the 0.12 video</a>+ </div>++ <h2>What you need</h2>+ <div class="table-wrap">+ <table class="cmp-table">+ <thead><tr><th></th><th class="us">Works</th><th>Recommended</th></tr></thead>+ <tbody>+ <tr><td>Windows or Linux + NVIDIA</td><td class="us">12 GB of VRAM (8 GB in beta)</td><td>16 GB or more</td></tr>+ <tr><td>Windows or Linux + AMD / Intel Arc</td><td class="us">12 GB of VRAM, Vulkan (beta)</td><td>16 GB or more</td></tr>+ <tr><td>Mac</td><td class="us">Apple Silicon, 16 GB, macOS 15</td><td>32 GB or more, Pro/Max/Ultra chip</td></tr>+ <tr><td>Disk</td><td class="us">15 GB free</td><td>25 GB free</td></tr>+ </tbody>+ </table>+ </div>++ <h2>Questions</h2>+ <div class="faq">+ <details><summary>Can AI split a flat JPG or PNG into separate layers?</summary><p>Yes. In Kokon's Layers lab you click each thing you want on its own layer; Kokon cuts it out and paints the background behind it, so the result is a stack of layers: the objects on top, the filled background at the bottom.</p></details>+ <details><summary>How do I convert an image to a PSD with layers for free?</summary><p>Open the image in Kokon, press Layers (lab), click the objects, press Split into layers, then Export PSD. Kokon is free and open source; the file is written on your computer.</p></details>+ <details><summary>What happens to the background behind an object I separate?</summary><p>Kokon repaints it with the image model, continuing the surfaces, light and perspective around it, so moving or hiding the object leaves no hole.</p></details>+ <details><summary>Does it work offline, without uploading my image?</summary><p>Yes. The cut-outs, the background fill and the PSD are all made on your own computer. After setup no internet is needed.</p></details>+ <details><summary>Can I edit the text in the image after the split?</summary><p>Text in the picture becomes an image layer you can move, hide or edit with a sentence, not editable type.</p></details>+ <details><summary>Which devices does it run on?</summary><p>Windows and Linux with a 12 GB graphics card (NVIDIA 8 GB in beta) and Apple Silicon Macs. The Layers lab is in the desktop app; it is off in the phone layout.</p></details>+ </div>+ <div class="cta-box">+ <h2>Try Kokon</h2>+ <p>Free and open source for Windows, macOS and Linux. One installer; Kokon sets up the model for your graphics card.</p>+ <div class="ctas" style="justify-content:flex-start">+ <a class="btn primary" href="https://github.com/GabrieleLozupone/kokon/releases/latest" rel="noopener" data-dl="windows"><svg fill="currentColor"><use href="#win"/></svg>Windows</a>+ <a class="btn ghost" href="https://github.com/GabrieleLozupone/kokon/releases/latest" rel="noopener" data-dl="mac"><svg fill="currentColor"><use href="#apple"/></svg>macOS <span class="tag">BETA</span></a>+ <a class="btn ghost" href="https://github.com/GabrieleLozupone/kokon/releases/latest" rel="noopener"><svg fill="currentColor"><use href="#linux"/></svg>Linux <span class="tag">BETA</span></a>+ <a class="btn ghost" href="https://github.com/GabrieleLozupone/kokon" rel="noopener"><svg fill="currentColor"><use href="#github"/></svg>Star on GitHub</a>+ </div>+ </div>+ </div>+</main>++<footer>+ <div class="wrap">+ <a class="logo" href="../"><svg><use href="#mark"/></svg>Kokon</a>+ <span>Designed and built by <a class="me" href="https://gabrielelozupone.github.io/" rel="author">Gabriele Lozupone</a>, with the help of Claude Code.</span>+ <nav aria-label="Links">+ <a href="../">Home</a>+ <a href="../local-ai-photo-editor/">Local AI photo editor</a>+ <a href="../offline-alternative-to-nano-banana/">Offline Nano Banana alternative</a>+ <a href="../comfyui-alternative/">Simple ComfyUI alternative</a>+ <a href="../remove-objects-from-photos-offline/">Remove objects offline</a>+ <a href="../image-to-layers/">Image to layers</a>+ <a href="../qwen-image-2-1-local/">Run Qwen-Image locally</a>+ <a href="../press/">Press kit</a>+ <a href="../support/">Support Kokon</a>+ <a href="https://github.com/GabrieleLozupone/kokon" rel="noopener">GitHub</a>+ </nav>+ <p class="legal">Nano Banana and Gemini are trademarks of Google; ChatGPT of OpenAI; Photoshop of Adobe; ComfyUI belongs to its authors. Qwen-Image is a model by the Qwen team at Alibaba. Kokon is an independent project and is not affiliated with any of them; the names are used only to compare.</p>+ </div>+</footer>+<dialog class="star-ask" id="star-ask" aria-labelledby="star-ask-title">+ <div>+ <svg class="mark"><use href="#mark"/></svg>+ <h2 id="star-ask-title">Your download has started</h2>+ <p>Kokon is free and open source, made by one person. If you like it, a star on GitHub helps other people find it and keeps the project going.</p>+ <form method="dialog" class="ctas">+ <a class="btn primary" id="star-ask-go" href="https://github.com/GabrieleLozupone/kokon" target="_blank" rel="noopener"><svg fill="currentColor"><use href="#github"/></svg>Star on GitHub</a>+ <button class="btn ghost">Maybe later</button>+ </form>+ </div>+</dialog>+<script src="/assets/layers-demo.js?v=1" defer></script>+<script src="/download.js" defer></script>+<script src="/nav.js" defer></script>+</body>+</html>diff --git a/site/index.html b/site/index.htmlindex e8862ee..0a8e128 100644--- a/site/index.html+++ b/site/index.html@@ -43,7 +43,7 @@"url": "https://kokon.app/","downloadUrl": "https://github.com/GabrieleLozupone/kokon/releases/latest","installUrl": "https://github.com/GabrieleLozupone/kokon/releases/latest",- "softwareVersion": "0.11.2",+ "softwareVersion": "0.12.0","releaseNotes": "https://github.com/GabrieleLozupone/kokon/releases/latest","softwareHelp": { "@type": "CreativeWork", "url": "https://kokon.app/prompt-guide/", "name": "Prompt guide for Qwen-Image 2.1" },"applicationCategory": "MultimediaApplication",@@ -51,15 +51,15 @@"operatingSystem": "Windows 10, Windows 11, macOS 15, Linux","processorRequirements": "NVIDIA, AMD or Intel GPU with 12 GB of VRAM (NVIDIA 8 GB in beta), or Apple Silicon","storageRequirements": "15 GB",- "fileSize": "53 MB",+ "fileSize": "136 MB","isAccessibleForFree": true,"license": "https://www.gnu.org/licenses/agpl-3.0.html","offers": { "@type": "Offer", "price": "0", "priceCurrency": "USD" },"screenshot": "https://kokon.app/assets/kokon-poster.jpg?v=3","image": "https://kokon.app/assets/og.png",- "featureList": ["Edit photos with a sentence", "Create images from text", "Prompt+: a few words become a detailed prompt, written by the model already loaded", "Your own Qwen-Image 2.1 LoRAs (beta)", "Paint an area to change only that area (inpainting)", "Remove, replace and add objects", "Expand images beyond their borders (outpainting)", "Before/after comparison and version history", "Use it from a phone, tablet or laptop in the browser, behind a password", "Docker image for NVIDIA GPUs", "Private mode", "MCP server and local HTTP API for AI agents", "Runs offline on NVIDIA, AMD, Intel and Apple Silicon"],+ "featureList": ["Edit photos with a sentence", "Create images from text", "Prompt+: a few words become a detailed prompt, written by the model already loaded", "Your own Qwen-Image 2.1 LoRAs (beta)", "Paint an area to change only that area (inpainting)", "Remove, replace and add objects", "Expand images beyond their borders (outpainting)", "Split an image into layers: each object cut out, the background filled in behind it (Layers lab, experimental)", "Export layers as a PSD for Photoshop, GIMP or Photopea", "Paint with colours: one colour per thing, say what goes there", "AI marking in every image: IPTC metadata and an invisible watermark", "Before/after comparison and version history", "Use it from a phone, tablet or laptop in the browser, behind a password", "Docker image for NVIDIA GPUs", "Private mode", "MCP server and local HTTP API for AI agents", "Runs offline on NVIDIA, AMD, Intel and Apple Silicon"],"author": { "@id": "https://gabrielelozupone.github.io/#me" },- "sameAs": ["https://github.com/GabrieleLozupone/kokon", "https://www.npmjs.com/package/kokon-mcp", "https://www.youtube.com/watch?v=y2NAJkZ4vH0"]+ "sameAs": ["https://github.com/GabrieleLozupone/kokon", "https://www.npmjs.com/package/kokon-mcp", "https://www.youtube.com/@glozupone", "https://www.youtube.com/watch?v=y2NAJkZ4vH0"]},{"@type": "Person",@@ -79,6 +79,25 @@"duration": "PT1M25S","embedUrl": "https://www.youtube.com/embed/y2NAJkZ4vH0"},+ {+ "@type": "VideoObject",+ "name": "Free AI photo editor on your PC: remove people, restore old photos, change colors",+ "description": "What Kokon can do, in examples made with it: remove photobombers, fix closed eyes, restore and colorize an old photo, add someone who is missing, try a haircut or a jacket, repaint a wall, turn a selfie into a headshot, put text on an invite.",+ "thumbnailUrl": "https://kokon.app/assets/img/videos/usecases.webp",+ "contentUrl": "https://kokon.app/assets/usecases.mp4?v=1",+ "uploadDate": "2026-11-04",+ "duration": "PT1M25S",+ "embedUrl": "https://www.youtube.com/embed/WDbHxkbpU4s"+ },+ {+ "@type": "VideoObject",+ "name": "Kokon 0.12: split any image into layers, move objects, export to PSD",+ "description": "The Layers lab: click each object, Kokon cuts it out onto its own layer and paints the background behind it; move, hide or edit each layer, blend it into the scene, export a PSD. Also painting with colours and Kokon's own Mac engine.",+ "thumbnailUrl": "https://kokon.app/assets/img/videos/update-0.12.webp",+ "uploadDate": "2026-11-04",+ "duration": "PT2M21S",+ "embedUrl": "https://www.youtube.com/embed/eGr9yYe7NVc"+ },{"@type": "WebSite","@id": "https://kokon.app/#site",@@ -92,7 +111,7 @@<link rel="preconnect" href="https://fonts.googleapis.com"><link rel="preconnect" href="https://fonts.gstatic.com" crossorigin><link href="https://fonts.googleapis.com/css2?family=Inter:opsz,[email protected],400..700&family=JetBrains+Mono:wght@400;500&display=swap" rel="stylesheet">-<link rel="stylesheet" href="assets/style.css?v=13">+<link rel="stylesheet" href="assets/style.css?v=14"><link rel="stylesheet" href="assets/devices.css?v=1"><script type="application/ld+json">{@@ -164,11 +183,16 @@}},{+ "@type": "Question",+ "name": "Can Kokon split an image into layers or make a PSD?",+ "acceptedAnswer": { "@type": "Answer", "text": "Yes, since 0.12, in the Layers lab (experimental). Click each thing you want on its own layer; Kokon cuts it out and paints the background behind it, so moving it leaves no hole. Then move, resize, hide or edit each layer, and save one image or a layered PSD for Photoshop, GIMP or Photopea. Text in the picture becomes an image layer, not editable type." }+ },+ {"@type": "Question","name": "Is it an alternative to Photoshop Generative Fill?","acceptedAnswer": {"@type": "Answer",- "text": "For many edits, yes: paint an area and describe what should be there, or remove, replace and add objects. It runs offline and costs nothing, but it is an AI editor, not a full layer-based image editor."+ "text": "For many edits, yes: paint an area and describe what should be there, or remove, replace and add objects. It runs offline and costs nothing. Since 0.12 it can also split an image into layers and export a PSD, but it is an AI editor, not a full layer-based editor like Photoshop."}},{@@ -282,6 +306,7 @@<li><a href="/self-hosted/"><b>Phone and server</b><span>Other devices, Linux, Docker, Tailscale</span></a></li><li><a href="/prompt-guide/"><b>Prompt guide</b><span>Write prompts that work</span></a></li><li><a href="/remove-objects-from-photos-offline/"><b>Remove objects</b><span>Step by step, offline</span></a></li>+ <li><a href="/image-to-layers/"><b>Image to layers <span class="new">New</span></b><span>Split a photo, export a PSD</span></a></li></ul></div></li><li class="dd"><button type="button" aria-expanded="false" aria-controls="dd-compare">Compare<svg class="caret" viewBox="0 0 12 12" aria-hidden="true"><path d="M3 4.5 6 7.5 9 4.5"/></svg></button><div class="dd-panel" id="dd-compare"><ul>@@ -305,25 +330,49 @@<main id="top"><section class="hero">- <div class="wrap">+ <div class="wall" id="wall" aria-hidden="true"></div>+ <div class="wrap hero-top"><span class="eyebrow rv"><b>Beta</b><span>A local alternative to Nano Banana<span class="eb-long"> and Generative Fill</span></span></span><h1 class="rv d1">Edit photos by describing the change.</h1>- <p class="lede rv d2">Kokon runs the best open image model on your own computer, set up for your GPU by one installer. Brush an area, type what you want, get the edit. Your photos stay with you, and you can edit as often as you like.</p><div class="ctas rv d3" id="download"><a class="btn primary" href="https://github.com/GabrieleLozupone/kokon/releases/latest" rel="noopener" data-dl="windows"><svg fill="currentColor"><use href="#win"/></svg>Download for Windows</a><a class="btn ghost" href="https://github.com/GabrieleLozupone/kokon/releases/latest" rel="noopener" data-dl="mac"><svg fill="currentColor"><use href="#apple"/></svg>Download for macOS <span class="tag">BETA</span></a><a class="btn ghost" href="https://github.com/GabrieleLozupone/kokon/releases/latest" rel="noopener"><svg fill="currentColor"><use href="#linux"/></svg>Download for Linux <span class="tag">BETA</span></a></div>- <p class="fine rv d3">Free and open source · Windows, Linux with a 12 GB GPU · Apple Silicon Macs · images for non-commercial use</p>+ <p class="fine rv d3">Free and open source · runs on your own GPU · images for non-commercial use</p></div><div class="wrap">- <div class="showreel rv">+ <div class="showreel rv" id="videos"><div class="glow"></div><div class="frame">- <video id="reel" src="assets/kokon.mp4?v=3" poster="assets/kokon-poster.jpg?v=3" muted loop playsinline preload="metadata" aria-label="Kokon in use: Prompt+ writing a prompt from a few words, painting an area to remove a scooter, and five more edits"></video>- <button class="sound" id="reel-btn" type="button" aria-label="Watch the film from the start with sound">🔊 Watch with sound</button>+ <div class="slide on">+ <video class="reel" src="assets/usecases.mp4?v=1" poster="assets/img/videos/usecases.webp" muted loop playsinline preload="metadata" aria-label="What you can do with Kokon: remove photobombers, fix closed eyes, restore and colorize an old photo, add someone who is missing, try a haircut or a jacket, repaint a wall, and more. All images are examples made with Kokon."></video>+ <button class="sound" type="button" aria-label="Watch the video from the start with sound">🔊 Watch with sound</button>+ </div>+ <div class="slide">+ <video class="reel" src="assets/kokon.mp4?v=3" poster="assets/kokon-poster.jpg?v=3" muted loop playsinline preload="none" aria-label="Kokon in use: Prompt+ writing a prompt from a few words, painting an area to remove a scooter, and five more edits"></video>+ <button class="sound" type="button" aria-label="Watch the film from the start with sound">🔊 Watch with sound</button>+ </div>+ <div class="slide"><button class="yt-facade" type="button" data-yt="5wqd9jus8C0" style="background-image:url(assets/img/videos/demo.webp)" aria-label="Play the video: from setup to the first edit (YouTube)"><span class="play"></span><span class="yt-note">Plays on YouTube</span></button></div>+ </div>+ <div class="vtabs" role="tablist" aria-label="Videos">+ <button class="vtab on" type="button" role="tab" aria-selected="true"><img src="assets/img/videos/usecases.webp" alt=""><div><b>What you can do with it</b><span>Everyday edits · 1:25</span></div><i class="bar"></i></button>+ <button class="vtab" type="button" role="tab" aria-selected="false"><img src="assets/kokon-poster.jpg?v=3" alt=""><div><b>Kokon in 85 seconds</b><span>The film · 1:25</span></div><i class="bar"></i></button>+ <button class="vtab" type="button" role="tab" aria-selected="false"><img src="assets/img/videos/demo.webp" alt=""><div><b>From setup to the first edit</b><span>Real-time demo · 0:58</span></div><i class="bar"></i></button></div>- <p class="reel-link"><a href="https://youtu.be/y2NAJkZ4vH0" rel="noopener">Watch in 4K on YouTube</a> · <a href="https://youtu.be/qGiGXasJJNI" rel="noopener">What's new in 0.10</a></p>+ </div>+ </div>+ </section>++ <section class="sec" id="layers">+ <div class="wrap center">+ <p class="kicker rv">New in 0.12 · Layers lab</p>+ <h2 class="rv d1" style="margin-left:auto;margin-right:auto">Any image, split into layers.</h2>+ <p class="sec-lede rv d2" style="margin-left:auto;margin-right:auto">Click the things you want. Each gets its own layer, with the background painted in behind.</p>+ <div class="layers-demo rv" data-layers-demo></div>+ <div class="layers-ctas rv">+ <a class="btn yt" href="https://youtu.be/eGr9yYe7NVc" rel="noopener"><svg><use href="#youtube"/></svg>Watch the 0.12 video</a>+ <a class="btn ghost" href="image-to-layers/">How to split an image into layers</a></div></div></section>@@ -488,7 +537,7 @@<article class="card b-small rv"><div class="icon"><svg viewBox="0 0 24 24"><path d="M21 15v4a2 2 0 0 1-2 2H5a2 2 0 0 1-2-2v-4M7 10l5 5 5-5M12 15V3"/></svg></div><h3>No Python. One installer.</h3>- <p>A small native app, about 55 MB. The models download and are checked on first run, with resume.</p>+ <p>A native app, about 140 to 170 MB. The models download and are checked on first run, with resume.</p><div class="vis"><div class="big-num">55<small>MB installer</small></div></div></article><article class="card b-wide rv d1">@@ -519,7 +568,7 @@<h2 class="rv d1">Run Qwen-Image 2.1 locally in three steps.</h2><p class="sec-lede rv d2">No Python environment, no ComfyUI workflows, no model files to hunt down. <a href="qwen-image-2-1-local/" style="text-decoration:underline">Read the full guide</a>.</p><div class="privacy">- <article class="card rv"><span class="step">1</span><h3>Download the installer</h3><p>A native app for Windows 10/11 (about 55 MB), macOS on Apple Silicon or Linux (.deb and .rpm, about 75 MB). On Windows it installs for your user, no admin rights needed.</p></article>+ <article class="card rv"><span class="step">1</span><h3>Download the installer</h3><p>A native app for Windows 10/11 (about 136 MB), macOS on Apple Silicon (about 162 MB) or Linux (.deb and .rpm, about 170 MB). On Windows it installs for your user, no admin rights needed.</p></article><article class="card rv d1"><span class="step">2</span><h3>Let it set up the model</h3><p>Kokon detects your GPU (NVIDIA, AMD, Intel or Apple Silicon), picks the right preset and downloads Qwen-Image 2.1 Turbo, with resume and SHA-256 checks.</p></article><article class="card rv d2"><span class="step">3</span><h3>Describe the change</h3><p>Drop a photo, write what you want and press Enter. Or paint an area first, and only that area changes.</p></article></div>@@ -634,7 +683,8 @@<details><summary>Does it work on AMD or Intel GPUs?</summary><p>Yes, in beta, through the Vulkan engine on Windows with 12 GB of VRAM or more. On NVIDIA it uses CUDA; on Apple Silicon Macs it uses Metal.</p></details><details><summary>My GPU has 8 GB. Will it work?</summary><p>Yes, in beta. Kokon makes images at about 1 megapixel on 8 GB cards; users report it works on an RTX 3060 Ti and an RTX 5050 Laptop (1 to 2 minutes per edit). 16 GB of system RAM is recommended. On a Mac, 16 GB of unified memory works but slowly; 32 GB is recommended.</p></details><details><summary>Can Claude Code or Cursor generate images with it?</summary><p>Yes. Kokon ships an MCP server: register it once and any MCP client can create, edit, inpaint, expand and upscale images on your GPU, for free and without an API key.</p></details>- <details><summary>Is it an alternative to Photoshop Generative Fill?</summary><p>For many edits, yes: paint an area and describe what should be there, or remove, replace and add objects. It runs offline and costs nothing, but it is an AI editor, not a full layer-based image editor.</p></details>+ <details><summary>Can Kokon split an image into layers or make a PSD?</summary><p>Yes, since 0.12, in the Layers lab (experimental). Click each thing you want on its own layer; Kokon cuts it out and paints the background behind it, so moving it leaves no hole. Then move, resize, hide or edit each layer, and save one image or a layered PSD for Photoshop, GIMP or Photopea. Text in the picture becomes an image layer, not editable type. More in <a href="image-to-layers/" style="text-decoration:underline">split an image into layers</a>.</p></details>+ <details><summary>Is it an alternative to Photoshop Generative Fill?</summary><p>For many edits, yes: paint an area and describe what should be there, or remove, replace and add objects. It runs offline and costs nothing. Since 0.12 it can also split an image into layers and export a PSD, but it is an AI editor, not a full layer-based editor like Photoshop.</p></details><details><summary>Can I use the images commercially?</summary><p>Not with the default model. The app is open source (AGPL-3.0), but Qwen-Image 2.1 is released under the Qwen Research License: research and evaluation only, no commercial use without a license from the Qwen team. The default upscaler (Superscale-SP) allows commercial use; the optional UltraSharp upscaler is non-commercial (CC BY-NC-SA 4.0).</p></details><details><summary>Windows says the app is unrecognized. Is it safe?</summary><p>The beta installer is not code-signed yet, so SmartScreen warns about it. Click "More info" and then "Run anyway". The code is public and every download the app makes is checked against a pinned hash.</p></details><details id="mac-open"><summary>macOS says Kokon cannot be opened. What do I do?</summary><p>The beta is not notarized by Apple yet, so macOS blocks it the first time:</p>@@ -684,6 +734,7 @@<a href="offline-alternative-to-nano-banana/">Offline Nano Banana alternative</a><a href="comfyui-alternative/">Simple ComfyUI alternative</a><a href="remove-objects-from-photos-offline/">Remove objects offline</a>+ <a href="image-to-layers/">Image to layers</a><a href="qwen-image-2-1-local/">Run Qwen-Image locally</a><a href="mcp-image-generation/">MCP image generation</a><a href="self-hosted/">Use it from your phone</a>@@ -774,12 +825,66 @@}), { threshold: 0.18 });document.querySelectorAll(".rv, #mask-demo, #cmp").forEach((el) => io.observe(el));- const reel = $("reel"), reelBtn = $("reel-btn");- const vio = new IntersectionObserver(([en]) => { if (en.isIntersecting) reel.play().catch(() => {}); else reel.pause(); }, { threshold: 0.3 });- vio.observe(reel);- // Browsers only autoplay muted, so the reel starts silent; the button restarts it with the- // voice-over and music.- reelBtn.addEventListener("click", () => { reel.currentTime = 0; reel.muted = false; reel.controls = true; reel.loop = false; reel.play().catch(() => {}); reelBtn.remove(); });+ // Hero wall: before/after pairs made by Kokon (small copies in assets/img/wall), three rows+ // scrolling in turn, each tile wiping to its edit with the request on it.+ const PAIRS = [+ ["street", "street-removed", "remove the red scooter"], ["uc01-before", "uc01-after", "remove the tourists"],+ ["cabin", "cabin-winter", "make it a snowy winter night"], ["uc03-before", "uc03-after", "restore this old photo"],+ ["bakery", "bakery-text", "change the sign to “KOKON CAFÉ”"], ["uc13-before", "uc13-after", "a copper bob haircut"],+ ["potter", "potter-watercolor", "turn it into a watercolor"], ["uc10-before", "uc10-after", "paint the wall sage green"],+ ["room", "room-sofa", "a green velvet sofa instead"], ["uc17-before", "uc17-after", "on a flamingo, surfing a wave"],+ ["sneaker", "sneaker-ad", "on wet black stone, advertising shot"], ["uc03-after", "uc03-color", "colorize it"],+ ["street-removed", "street-cat", "add a ginger cat"], ["uc04-before", "uc04-after", "add grandma from this photo"],+ ["lighthouse-framed", "lighthouse-wide", "expand: continue the coastline"], ["uc16-before", "uc16-after", "a professional headshot"],+ ["uc12-before", "uc12-after", "tidy up the garden"], ["uc11-before", "uc11-after", "a leather sofa"], ["uc05-before", "uc05-after", "tidy up the room"],+ ];+ const wall = $("wall");+ if (wall) {+ const w = (n) => `assets/img/wall/${n}.webp`;+ const tile = ([a, b, p], k) => `<div class="wall-tile" style="--d:${-((k * 3.7) % 10).toFixed(1)}s"><img src="${w(a)}" alt="" loading="lazy" decoding="async"><div class="aw"><img src="${w(b)}" alt="" loading="lazy" decoding="async"></div><span class="chip">${p}</span></div>`;+ wall.innerHTML = `<div class="wall-plane">${[0, 1, 2].map((r) => {+ const row = PAIRS.slice(r * 6).concat(PAIRS.slice(0, r * 6));+ const html = row.map((p, i) => tile(p, i + r * 5)).join("");+ return `<div class="wall-row${r % 2 ? " r" : ""}" style="animation-duration:${110 + r * 18}s">${html}${html}</div>`;+ }).join("")}</div>`;+ }++ // Videos: the film plays here; the others are YouTube, loaded only when played.+ const slides = [...document.querySelectorAll("#videos .slide")], vtabs = [...document.querySelectorAll("#videos .vtab")];+ let vcur = 0;+ const showVideo = (i) => {+ vcur = (i + slides.length) % slides.length;+ slides.forEach((s, j) => {+ s.classList.toggle("on", j === vcur);+ if (j !== vcur) { const f = s.querySelector("iframe"); if (f) f.replaceWith(f._facade); }+ });+ vtabs.forEach((b, j) => { b.classList.toggle("on", j === vcur); b.setAttribute("aria-selected", j === vcur); });+ playShown();+ };+ vtabs.forEach((b, j) => b.addEventListener("click", () => showVideo(j)));+ document.querySelectorAll(".yt-facade").forEach((f) => f.addEventListener("click", () => {+ const ifr = document.createElement("iframe");+ ifr.src = `https://www.youtube-nocookie.com/embed/${f.dataset.yt}?autoplay=1&rel=0`;+ ifr.title = f.getAttribute("aria-label").replace("Play the video: ", "");+ ifr.allow = "autoplay; encrypted-media; picture-in-picture; fullscreen";+ ifr.allowFullscreen = true;+ ifr._facade = f;+ f.replaceWith(ifr);+ }));++ // The local videos (use cases first, then the film) play muted while their slide is shown and the+ // carousel is on screen; their button restarts the video with the voice-over and music.+ let videosSeen = false;+ const playShown = () => slides.forEach((s, j) => { const v = s.querySelector("video.reel"); if (!v) return; if (j === vcur && videosSeen) v.play().catch(() => {}); else v.pause(); });+ new IntersectionObserver(([en]) => { videosSeen = en.isIntersecting; playShown(); }, { threshold: 0.3 }).observe(document.querySelector("#videos .frame"));+ slides.forEach((s, j) => {+ const vv = s.querySelector("video.reel"), bar = vtabs[j] && vtabs[j].querySelector(".bar");+ if (vv && bar) vv.addEventListener("timeupdate", () => { bar.style.width = `${(vv.currentTime / (vv.duration || 1)) * 100}%`; });+ });+ slides.forEach((s) => {+ const v = s.querySelector("video.reel"), btn = s.querySelector(".sound");+ if (v && btn) btn.addEventListener("click", () => { v.currentTime = 0; v.muted = false; v.controls = true; v.loop = false; v.play().catch(() => {}); btn.remove(); });+ });// Other devices: the scene is drawn at a fixed size and scaled to the column; on a narrow// screen it drops the laptop and moves the parts (data-narrow).@@ -819,6 +924,7 @@</form></div></dialog>+<script src="/assets/layers-demo.js?v=1" defer></script><script src="/download.js" defer></script><script src="/nav.js" defer></script></body>diff --git a/site/llms.txt b/site/llms.txtindex 7d0ffbd..b4aeec1 100644--- a/site/llms.txt+++ b/site/llms.txt@@ -2,14 +2,15 @@> Kokon is a free, open-source desktop app for editing and creating photos with AI on your own computer, a free, local alternative to Nano Banana and ChatGPT image editing: photos stay on the user's computer and there is no daily limit. Kokon picks and sets up the model for you. You describe the change in plain words ("remove the scooter", "make it a snowy night") and Qwen-Image 2.1 does it locally on your GPU. There is nothing to set up: one installer detects the hardware, downloads and verifies the right model, and picks the resolution and the VRAM/RAM split for every job. Nothing is uploaded.-Key facts (version 0.11.2, October 2026):+Key facts (version 0.12.0, November 2026):- Model: Qwen-Image 2.1 Turbo (official 8-step version, 4-bit GGUF) with Qwen3-VL 8B as the text encoder, run through stable-diffusion.cpp. About 12 GB of models, downloaded once and checked with SHA-256.- Platforms: Windows 10/11 with an NVIDIA GPU (12 GB VRAM or more, CUDA; 8 GB in beta, about 1-megapixel images). AMD and Intel Arc (12 GB+) through Vulkan, beta. Apple Silicon Macs on macOS 15 or newer (Metal), beta: 16 GB works slowly, 32 GB recommended. Linux x86-64 (.deb for Ubuntu/Debian/Mint, .rpm for Fedora; CUDA or Vulkan), beta.-- Speed: RTX 5070 Ti about 24 s per edit at 1.5K and 36 s for a new image at native 2K; RTX 5070 12 GB about 12 s per 1-megapixel image; M2 Pro about 4 minutes per image.+- Speed: RTX 5070 Ti about 24 s per edit at 1.5K and 36 s for a new image at native 2K; RTX 5070 12 GB about 12 s per 1-megapixel image; M2 Pro 16 GB about 65 s per new 1-megapixel image with Kokon's own Mac engine (about 106 s per edit).- Features: Prompt+ (a few words in any language become a detailed prompt, written by the Qwen3-VL text encoder already loaded, inside Kokon's own build of stable-diffusion.cpp, so no second model), your own Qwen-Image 2.1 LoRAs (beta, tested on Windows), create from text, new images from reference photos (a person, a product, a style), edit with a sentence, paint an area and Change/Remove/Replace/Add only there, combine several images ("image 1", "image 2"), expand beyond the borders (outpainting), AI upscaling, versions with undo and before/after compare, private mode, works offline after setup.- For AI agents: built-in MCP server (generate_image, edit_image, inpaint_image, outpaint_image, upscale_image, ...) and a local HTTP API. `npx -y kokon-mcp`; listed in the official MCP Registry as io.github.GabrieleLozupone/kokon.- License: the app is AGPL-3.0. The Qwen-Image 2.1 weights are under the Qwen Research License, which does not allow commercial use.+- Layers lab (0.12, experimental): click the objects of any image; Kokon cuts each onto its own layer and paints the background behind it; move, hide, edit or expand each layer, blend objects into the scene, export a layered PSD. Also in 0.12: paint with colours (one colour per thing, say what goes there), Kokon's own Mac engine as the default on Apple Silicon, AI marking in every image (IPTC metadata plus an invisible watermark).- Other devices (0.11): the interface in a browser on a phone, tablet or laptop on the same network, behind a password (off by default); headless `kokon-service` on Linux; Docker image ghcr.io/gabrielelozupone/kokon for NVIDIA GPUs; Tailscale or another VPN for access from outside the home, never an open port.- Not yet: a Docker image for AMD/Intel (Vulkan); code signing (Windows SmartScreen and macOS Gatekeeper warn on first run).- Made by Gabriele Lozupone, an AI researcher. Independent project, not affiliated with Alibaba or the Qwen team.@@ -21,6 +22,7 @@ Key facts (version 0.11.2, October 2026):- [Offline alternative to Nano Banana](https://kokon.app/offline-alternative-to-nano-banana/): Kokon vs Nano Banana / ChatGPT image editing: what's the same, what's different, when the cloud is still better- [Simple ComfyUI alternative](https://kokon.app/comfyui-alternative/): what Kokon decides for you (model, files, steps, VRAM), reusing ComfyUI model files, when to stay with ComfyUI- [Remove objects from photos offline](https://kokon.app/remove-objects-from-photos-offline/): step-by-step object and people removal on your own GPU+- [Split an image into layers](https://kokon.app/image-to-layers/): image to layers and a layered PSD, offline: how it works, the filled background, PSD export, FAQ- [Press kit](https://kokon.app/press/): one-line and short descriptions, key facts, videos, images, logos- [Run Qwen-Image 2.1 locally](https://kokon.app/qwen-image-2-1-local/): step-by-step setup on Windows, Mac and Linux without Python or ComfyUI, speed, troubleshooting- [Prompt guide](https://kokon.app/prompt-guide/): how to write prompts for Qwen-Image 2.1, with vague vs clear examples (negative prompts, text in images, edit prompts)diff --git a/site/local-ai-photo-editor/index.html b/site/local-ai-photo-editor/index.htmlindex f2a5035..8e98ed6 100644--- a/site/local-ai-photo-editor/index.html+++ b/site/local-ai-photo-editor/index.html@@ -22,7 +22,7 @@<link rel="preconnect" href="https://fonts.googleapis.com"><link rel="preconnect" href="https://fonts.gstatic.com" crossorigin><link href="https://fonts.googleapis.com/css2?family=Inter:opsz,[email protected],400..700&family=JetBrains+Mono:wght@400;500&display=swap" rel="stylesheet">-<link rel="stylesheet" href="../assets/style.css?v=13">+<link rel="stylesheet" href="../assets/style.css?v=14"><script type="application/ld+json">{"@context": "https://schema.org",@@ -126,7 +126,8 @@<div class="dd-panel" id="dd-product"><ul><li><a href="/#examples"><b>Examples</b><span>Real edits, before and after</span></a></li><li><a href="/#features"><b>Features</b><span>Edit, paint, combine, expand, upscale</span></a></li>- <li><a href="/#devices"><b>On your phone <span class="new">New</span></b><span>Use your GPU from any device</span></a></li>+ <li><a href="/#layers"><b>Layers lab <span class="new">New</span></b><span>Split an image, move the pieces</span></a></li>+ <li><a href="/#devices"><b>On your phone</b><span>Use your GPU from any device</span></a></li><li><a href="/mcp-image-generation/"><b>For AI agents</b><span>MCP server and HTTP API</span></a></li><li><a href="/#requirements"><b>Requirements</b><span>GPUs, Macs and disk space</span></a></li></ul></div></li>@@ -136,6 +137,7 @@<li><a href="/self-hosted/"><b>Phone and server</b><span>Other devices, Linux, Docker, Tailscale</span></a></li><li><a href="/prompt-guide/"><b>Prompt guide</b><span>Write prompts that work</span></a></li><li><a href="/remove-objects-from-photos-offline/"><b>Remove objects</b><span>Step by step, offline</span></a></li>+ <li><a href="/image-to-layers/"><b>Image to layers <span class="new">New</span></b><span>Split a photo, export a PSD</span></a></li></ul></div></li><li class="dd"><button type="button" aria-expanded="false" aria-controls="dd-compare">Compare<svg class="caret" viewBox="0 0 12 12" aria-hidden="true"><path d="M3 4.5 6 7.5 9 4.5"/></svg></button><div class="dd-panel" id="dd-compare"><ul>@@ -217,11 +219,11 @@</tbody></table></div>- <p>Speed with the default model: on an RTX 5070 Ti, about 24 seconds for an edit at 1.5K and about 36 seconds for a new image at native 2K. 12 GB cards are a little slower, 8 GB cards take 1 to 2 minutes per edit, and an M2 Pro Mac takes about 4 minutes per image.</p>+ <p>Speed with the default model: on an RTX 5070 Ti, about 24 seconds for an edit at 1.5K and about 36 seconds for a new image at native 2K. 12 GB cards are a little slower, 8 GB cards take 1 to 2 minutes per edit, and an M2 Pro Mac takes about a minute for a new 1-megapixel image (about 106 s for an edit) with Kokon's own Mac engine.</p><h2>How to start</h2><ol>- <li><strong>Download Kokon</strong> for Windows, macOS or Linux (about 55 MB).</li>+ <li><strong>Download Kokon</strong> for Windows, macOS or Linux (about 136 MB on Windows, 162 MB on a Mac, 170 MB on Linux).</li><li><strong>Open it.</strong> It detects your graphics card, recommends a preset and downloads the model once (about 12 GB, checked and resumable).</li><li><strong>Drop a photo and describe the change.</strong> Or paint an area first, and only that area changes. If you only have a short idea, turn on <em>Prompt+</em> and it writes the detailed prompt for you, in any language.</li></ol>@@ -230,7 +232,7 @@<ul><li><strong>You need a graphics card</strong> with 12 GB of memory (8 GB in beta) or an Apple Silicon Mac. On a laptop with integrated graphics, a cloud tool is the better choice.</li><li><strong>Non-commercial images:</strong> the app is open source, but the default model (Qwen-Image 2.1) is under the Qwen Research License, which doesn't allow commercial use.</li>- <li><strong>Not a layer editor:</strong> Kokon is for AI edits. For layers, curves and retouching, keep your usual editor and use Kokon next to it.</li>+ <li><strong>Not a full layer editor:</strong> Kokon is for AI edits. Since 0.12 it can <a href="../image-to-layers/" style="text-decoration:underline">split an image into layers</a> and export a PSD, but for curves, masks and retouching keep your usual editor and use Kokon next to it.</li></ul><h2>Questions</h2>@@ -264,6 +266,7 @@<a href="../offline-alternative-to-nano-banana/">Offline Nano Banana alternative</a><a href="../comfyui-alternative/">Simple ComfyUI alternative</a><a href="../remove-objects-from-photos-offline/">Remove objects offline</a>+ <a href="../image-to-layers/">Image to layers</a><a href="../qwen-image-2-1-local/">Run Qwen-Image locally</a><a href="../press/">Press kit</a><a href="../support/">Support Kokon</a>diff --git a/site/mcp-image-generation/index.html b/site/mcp-image-generation/index.htmlindex 59f2fff..44350bd 100644--- a/site/mcp-image-generation/index.html+++ b/site/mcp-image-generation/index.html@@ -22,7 +22,7 @@<link rel="preconnect" href="https://fonts.googleapis.com"><link rel="preconnect" href="https://fonts.gstatic.com" crossorigin><link href="https://fonts.googleapis.com/css2?family=Inter:opsz,[email protected],400..700&family=JetBrains+Mono:wght@400;500&display=swap" rel="stylesheet">-<link rel="stylesheet" href="../assets/style.css?v=13">+<link rel="stylesheet" href="../assets/style.css?v=14"><script type="application/ld+json">{"@context": "https://schema.org",@@ -73,7 +73,8 @@<div class="dd-panel" id="dd-product"><ul><li><a href="/#examples"><b>Examples</b><span>Real edits, before and after</span></a></li><li><a href="/#features"><b>Features</b><span>Edit, paint, combine, expand, upscale</span></a></li>- <li><a href="/#devices"><b>On your phone <span class="new">New</span></b><span>Use your GPU from any device</span></a></li>+ <li><a href="/#layers"><b>Layers lab <span class="new">New</span></b><span>Split an image, move the pieces</span></a></li>+ <li><a href="/#devices"><b>On your phone</b><span>Use your GPU from any device</span></a></li><li><a href="/mcp-image-generation/"><b>For AI agents</b><span>MCP server and HTTP API</span></a></li><li><a href="/#requirements"><b>Requirements</b><span>GPUs, Macs and disk space</span></a></li></ul></div></li>@@ -83,6 +84,7 @@<li><a href="/self-hosted/"><b>Phone and server</b><span>Other devices, Linux, Docker, Tailscale</span></a></li><li><a href="/prompt-guide/"><b>Prompt guide</b><span>Write prompts that work</span></a></li><li><a href="/remove-objects-from-photos-offline/"><b>Remove objects</b><span>Step by step, offline</span></a></li>+ <li><a href="/image-to-layers/"><b>Image to layers <span class="new">New</span></b><span>Split a photo, export a PSD</span></a></li></ul></div></li><li class="dd"><button type="button" aria-expanded="false" aria-controls="dd-compare">Compare<svg class="caret" viewBox="0 0 12 12" aria-hidden="true"><path d="M3 4.5 6 7.5 9 4.5"/></svg></button><div class="dd-panel" id="dd-compare"><ul>diff --git a/site/offline-alternative-to-nano-banana/index.html b/site/offline-alternative-to-nano-banana/index.htmlindex 68b7a68..ae2700e 100644--- a/site/offline-alternative-to-nano-banana/index.html+++ b/site/offline-alternative-to-nano-banana/index.html@@ -22,7 +22,7 @@<link rel="preconnect" href="https://fonts.googleapis.com"><link rel="preconnect" href="https://fonts.gstatic.com" crossorigin><link href="https://fonts.googleapis.com/css2?family=Inter:opsz,[email protected],400..700&family=JetBrains+Mono:wght@400;500&display=swap" rel="stylesheet">-<link rel="stylesheet" href="../assets/style.css?v=13">+<link rel="stylesheet" href="../assets/style.css?v=14"><script type="application/ld+json">{"@context": "https://schema.org",@@ -126,7 +126,8 @@<div class="dd-panel" id="dd-product"><ul><li><a href="/#examples"><b>Examples</b><span>Real edits, before and after</span></a></li><li><a href="/#features"><b>Features</b><span>Edit, paint, combine, expand, upscale</span></a></li>- <li><a href="/#devices"><b>On your phone <span class="new">New</span></b><span>Use your GPU from any device</span></a></li>+ <li><a href="/#layers"><b>Layers lab <span class="new">New</span></b><span>Split an image, move the pieces</span></a></li>+ <li><a href="/#devices"><b>On your phone</b><span>Use your GPU from any device</span></a></li><li><a href="/mcp-image-generation/"><b>For AI agents</b><span>MCP server and HTTP API</span></a></li><li><a href="/#requirements"><b>Requirements</b><span>GPUs, Macs and disk space</span></a></li></ul></div></li>@@ -136,6 +137,7 @@<li><a href="/self-hosted/"><b>Phone and server</b><span>Other devices, Linux, Docker, Tailscale</span></a></li><li><a href="/prompt-guide/"><b>Prompt guide</b><span>Write prompts that work</span></a></li><li><a href="/remove-objects-from-photos-offline/"><b>Remove objects</b><span>Step by step, offline</span></a></li>+ <li><a href="/image-to-layers/"><b>Image to layers <span class="new">New</span></b><span>Split a photo, export a PSD</span></a></li></ul></div></li><li class="dd"><button type="button" aria-expanded="false" aria-controls="dd-compare">Compare<svg class="caret" viewBox="0 0 12 12" aria-hidden="true"><path d="M3 4.5 6 7.5 9 4.5"/></svg></button><div class="dd-panel" id="dd-compare"><ul>@@ -232,7 +234,7 @@</tbody></table></div>- <p>Speed with the default model: on an RTX 5070 Ti, about 24 seconds for an edit at 1.5K and about 36 seconds for a new image at native 2K. 12 GB cards are a little slower, 8 GB cards take 1 to 2 minutes per edit, and an M2 Pro Mac takes about 4 minutes per image.</p>+ <p>Speed with the default model: on an RTX 5070 Ti, about 24 seconds for an edit at 1.5K and about 36 seconds for a new image at native 2K. 12 GB cards are a little slower, 8 GB cards take 1 to 2 minutes per edit, and an M2 Pro Mac takes about a minute for a new 1-megapixel image (about 106 s for an edit) with Kokon's own Mac engine.</p><h2>Questions</h2><div class="faq">@@ -265,6 +267,7 @@<a href="../offline-alternative-to-nano-banana/">Offline Nano Banana alternative</a><a href="../comfyui-alternative/">Simple ComfyUI alternative</a><a href="../remove-objects-from-photos-offline/">Remove objects offline</a>+ <a href="../image-to-layers/">Image to layers</a><a href="../qwen-image-2-1-local/">Run Qwen-Image locally</a><a href="../press/">Press kit</a><a href="../support/">Support Kokon</a>diff --git a/site/press/index.html b/site/press/index.htmlindex 829ef8b..4273c82 100644--- a/site/press/index.html+++ b/site/press/index.html@@ -22,7 +22,7 @@<link rel="preconnect" href="https://fonts.googleapis.com"><link rel="preconnect" href="https://fonts.gstatic.com" crossorigin><link href="https://fonts.googleapis.com/css2?family=Inter:opsz,[email protected],400..700&family=JetBrains+Mono:wght@400;500&display=swap" rel="stylesheet">-<link rel="stylesheet" href="../assets/style.css?v=13">+<link rel="stylesheet" href="../assets/style.css?v=14"><script type="application/ld+json">{"@context": "https://schema.org",@@ -89,7 +89,8 @@<div class="dd-panel" id="dd-product"><ul><li><a href="/#examples"><b>Examples</b><span>Real edits, before and after</span></a></li><li><a href="/#features"><b>Features</b><span>Edit, paint, combine, expand, upscale</span></a></li>- <li><a href="/#devices"><b>On your phone <span class="new">New</span></b><span>Use your GPU from any device</span></a></li>+ <li><a href="/#layers"><b>Layers lab <span class="new">New</span></b><span>Split an image, move the pieces</span></a></li>+ <li><a href="/#devices"><b>On your phone</b><span>Use your GPU from any device</span></a></li><li><a href="/mcp-image-generation/"><b>For AI agents</b><span>MCP server and HTTP API</span></a></li><li><a href="/#requirements"><b>Requirements</b><span>GPUs, Macs and disk space</span></a></li></ul></div></li>@@ -99,6 +100,7 @@<li><a href="/self-hosted/"><b>Phone and server</b><span>Other devices, Linux, Docker, Tailscale</span></a></li><li><a href="/prompt-guide/"><b>Prompt guide</b><span>Write prompts that work</span></a></li><li><a href="/remove-objects-from-photos-offline/"><b>Remove objects</b><span>Step by step, offline</span></a></li>+ <li><a href="/image-to-layers/"><b>Image to layers <span class="new">New</span></b><span>Split a photo, export a PSD</span></a></li></ul></div></li><li class="dd"><button type="button" aria-expanded="false" aria-controls="dd-compare">Compare<svg class="caret" viewBox="0 0 12 12" aria-hidden="true"><path d="M3 4.5 6 7.5 9 4.5"/></svg></button><div class="dd-panel" id="dd-compare"><ul>@@ -138,7 +140,7 @@<li><strong>Platforms:</strong> Windows 10/11, macOS 15+ on Apple Silicon (beta), Linux x86-64 .deb and .rpm (beta).</li><li><strong>Hardware:</strong> NVIDIA with 12 GB of VRAM (8 GB in beta); AMD and Intel Arc with 12 GB through Vulkan (beta); Apple Silicon with 16 GB or more (32 GB recommended).</li><li><strong>Model:</strong> Qwen-Image 2.1 Turbo (8 steps), run through Kokon's own build of stable-diffusion.cpp. About 12 GB, downloaded once.</li>- <li><strong>Speed:</strong> about 24 s per edit at 1.5K and 36 s per new image at native 2K on an RTX 5070 Ti; from first launch to the first edit in under 4 minutes there, including the 12 GB download on a fast connection (see the demo). Macs are slower: about 4 minutes for 1 MP on an M2 Pro 16 GB, or about 65 s for a new 1 MP image (about 106 s for an edit) with Kokon's experimental Mac engine (Settings → Performance).</li>+ <li><strong>Speed:</strong> about 24 s per edit at 1.5K and 36 s per new image at native 2K on an RTX 5070 Ti; from first launch to the first edit in under 4 minutes there, including the 12 GB download on a fast connection (see the demo). On an M2 Pro 16 GB Mac, Kokon's own Mac engine (the default since 0.12) makes a new 1 MP image in about 65 s (about 106 s for an edit).</li><li><strong>Features:</strong> edit with a sentence; paint an area and Change, Remove, Replace or Add; combine photos; expand beyond the borders; create from text; Prompt+ (writes the detailed prompt from a few words); your own LoRAs (beta); 2× and 4K upscaling; versions with before/after; private mode; use from your phone or another device in the browser; a server mode for Linux and Docker; MCP server and local API.</li><li><strong>Made by:</strong> Gabriele Lozupone, an AI researcher in Italy, with the help of Claude Code. Independent project.</li><li><strong>Released:</strong> October 2026. Current version on the <a href="https://github.com/GabrieleLozupone/kokon/releases/latest" rel="noopener">releases page</a>.</li>@@ -147,6 +149,9 @@<h2>Videos</h2><ul><li><a href="https://youtu.be/5wqd9jus8C0" rel="noopener"><strong>Demo: from setup to the first edit</strong></a> (58 s, real time on screen, RTX 5070 Ti)</li>+ <li><a href="https://youtu.be/WDbHxkbpU4s" rel="noopener"><strong>What you can do with Kokon</strong></a> (85 s, 4K: remove people, restore and colorize, try-ons, home, text)</li>+ <li><a href="https://youtu.be/eGr9yYe7NVc" rel="noopener">What's new in 0.12</a> (2:21, 4K: Layers lab, PSD, paint with colours, Mac engine)</li>+ <li><a href="https://youtu.be/XnC_7P2_cww" rel="noopener">What's new in 0.11</a> (77 s: Kokon on your phone)</li><li><a href="https://youtu.be/y2NAJkZ4vH0" rel="noopener">Kokon 0.10 film</a> (85 s, 4K)</li><li><a href="https://youtu.be/qGiGXasJJNI" rel="noopener">What's new in 0.10</a> (42 s, 4K)</li><li><a href="../assets/kokon.mp4">Download the film as MP4</a></li>@@ -194,6 +199,7 @@<a href="../offline-alternative-to-nano-banana/">Offline Nano Banana alternative</a><a href="../comfyui-alternative/">Simple ComfyUI alternative</a><a href="../remove-objects-from-photos-offline/">Remove objects offline</a>+ <a href="../image-to-layers/">Image to layers</a><a href="../qwen-image-2-1-local/">Run Qwen-Image locally</a><a href="../press/">Press kit</a><a href="../support/">Support Kokon</a>diff --git a/site/prompt-guide/index.html b/site/prompt-guide/index.htmlindex 6a5c3fe..c120a4a 100644--- a/site/prompt-guide/index.html+++ b/site/prompt-guide/index.html@@ -26,7 +26,7 @@<link rel="preconnect" href="https://fonts.googleapis.com"><link rel="preconnect" href="https://fonts.gstatic.com" crossorigin><link href="https://fonts.googleapis.com/css2?family=Inter:opsz,[email protected],400..700&family=JetBrains+Mono:wght@400;500&display=swap" rel="stylesheet">-<link rel="stylesheet" href="../assets/style.css?v=13">+<link rel="stylesheet" href="../assets/style.css?v=14"><script type="application/ld+json">{"@context": "https://schema.org",@@ -109,7 +109,8 @@<div class="dd-panel" id="dd-product"><ul><li><a href="/#examples"><b>Examples</b><span>Real edits, before and after</span></a></li><li><a href="/#features"><b>Features</b><span>Edit, paint, combine, expand, upscale</span></a></li>- <li><a href="/#devices"><b>On your phone <span class="new">New</span></b><span>Use your GPU from any device</span></a></li>+ <li><a href="/#layers"><b>Layers lab <span class="new">New</span></b><span>Split an image, move the pieces</span></a></li>+ <li><a href="/#devices"><b>On your phone</b><span>Use your GPU from any device</span></a></li><li><a href="/mcp-image-generation/"><b>For AI agents</b><span>MCP server and HTTP API</span></a></li><li><a href="/#requirements"><b>Requirements</b><span>GPUs, Macs and disk space</span></a></li></ul></div></li>@@ -119,6 +120,7 @@<li><a href="/self-hosted/"><b>Phone and server</b><span>Other devices, Linux, Docker, Tailscale</span></a></li><li><a href="/prompt-guide/"><b>Prompt guide</b><span>Write prompts that work</span></a></li><li><a href="/remove-objects-from-photos-offline/"><b>Remove objects</b><span>Step by step, offline</span></a></li>+ <li><a href="/image-to-layers/"><b>Image to layers <span class="new">New</span></b><span>Split a photo, export a PSD</span></a></li></ul></div></li><li class="dd"><button type="button" aria-expanded="false" aria-controls="dd-compare">Compare<svg class="caret" viewBox="0 0 12 12" aria-hidden="true"><path d="M3 4.5 6 7.5 9 4.5"/></svg></button><div class="dd-panel" id="dd-compare"><ul>diff --git a/site/qwen-image-2-1-local/index.html b/site/qwen-image-2-1-local/index.htmlindex 5973422..3bb7209 100644--- a/site/qwen-image-2-1-local/index.html+++ b/site/qwen-image-2-1-local/index.html@@ -22,7 +22,7 @@<link rel="preconnect" href="https://fonts.googleapis.com"><link rel="preconnect" href="https://fonts.gstatic.com" crossorigin><link href="https://fonts.googleapis.com/css2?family=Inter:opsz,[email protected],400..700&family=JetBrains+Mono:wght@400;500&display=swap" rel="stylesheet">-<link rel="stylesheet" href="../assets/style.css?v=13">+<link rel="stylesheet" href="../assets/style.css?v=14"><script type="application/ld+json">{"@context": "https://schema.org",@@ -51,7 +51,7 @@{ "@type": "Question", "name": "How much VRAM does Qwen-Image 2.1 need?", "acceptedAnswer": { "@type": "Answer", "text": "With Kokon, 12 GB of VRAM is enough (tested on an RTX 5070 12 GB), and NVIDIA cards with 8 GB work in beta at about 1 megapixel. 16 GB or more is more comfortable. Apple Silicon Macs work with 16 GB of unified memory, slowly; 32 GB is recommended." } },{ "@type": "Question", "name": "Can I run Qwen-Image 2.1 without ComfyUI or Python?", "acceptedAnswer": { "@type": "Answer", "text": "Yes. Kokon runs it through stable-diffusion.cpp, a native C++ engine, and ships as one installer. It downloads and checks the model on first run." } },{ "@type": "Question", "name": "Does Qwen-Image 2.1 run on AMD or Intel GPUs?", "acceptedAnswer": { "@type": "Answer", "text": "Yes, through Vulkan on Windows (beta in Kokon), with 12 GB of VRAM or more." } },- { "@type": "Question", "name": "How fast is Qwen-Image 2.1 locally?", "acceptedAnswer": { "@type": "Answer", "text": "With the official 8-step Turbo model on an RTX 5070 Ti: about 24 s for an edit at 1.5K and about 36 s for a new image at native 2K. An M2 Pro Mac takes about 4 minutes for a 1-megapixel image." } }+ { "@type": "Question", "name": "How fast is Qwen-Image 2.1 locally?", "acceptedAnswer": { "@type": "Answer", "text": "With the official 8-step Turbo model on an RTX 5070 Ti: about 24 s for an edit at 1.5K and about 36 s for a new image at native 2K. An M2 Pro Mac takes about 65 seconds for a new 1-megapixel image with Kokon's own Mac engine (about 106 s for an edit)." } }]}]@@ -74,7 +74,8 @@<div class="dd-panel" id="dd-product"><ul><li><a href="/#examples"><b>Examples</b><span>Real edits, before and after</span></a></li><li><a href="/#features"><b>Features</b><span>Edit, paint, combine, expand, upscale</span></a></li>- <li><a href="/#devices"><b>On your phone <span class="new">New</span></b><span>Use your GPU from any device</span></a></li>+ <li><a href="/#layers"><b>Layers lab <span class="new">New</span></b><span>Split an image, move the pieces</span></a></li>+ <li><a href="/#devices"><b>On your phone</b><span>Use your GPU from any device</span></a></li><li><a href="/mcp-image-generation/"><b>For AI agents</b><span>MCP server and HTTP API</span></a></li><li><a href="/#requirements"><b>Requirements</b><span>GPUs, Macs and disk space</span></a></li></ul></div></li>@@ -84,6 +85,7 @@<li><a href="/self-hosted/"><b>Phone and server</b><span>Other devices, Linux, Docker, Tailscale</span></a></li><li><a href="/prompt-guide/"><b>Prompt guide</b><span>Write prompts that work</span></a></li><li><a href="/remove-objects-from-photos-offline/"><b>Remove objects</b><span>Step by step, offline</span></a></li>+ <li><a href="/image-to-layers/"><b>Image to layers <span class="new">New</span></b><span>Split a photo, export a PSD</span></a></li></ul></div></li><li class="dd"><button type="button" aria-expanded="false" aria-controls="dd-compare">Compare<svg class="caret" viewBox="0 0 12 12" aria-hidden="true"><path d="M3 4.5 6 7.5 9 4.5"/></svg></button><div class="dd-panel" id="dd-compare"><ul>@@ -138,7 +140,7 @@<h2>Step 1: Install Kokon</h2><p>Download the latest release from <a href="https://github.com/GabrieleLozupone/kokon/releases/latest" rel="noopener">GitHub</a>:</p><ul>- <li><strong>Windows:</strong> <code>Kokon_<version>_x64-setup.exe</code> (about 55 MB). It installs for your user only, no administrator rights. The beta is not code-signed yet, so if SmartScreen warns, click <em>More info → Run anyway</em>.</li>+ <li><strong>Windows:</strong> <code>Kokon_<version>_x64-setup.exe</code> (about 136 MB). It installs for your user only, no administrator rights. The beta is not code-signed yet, so if SmartScreen warns, click <em>More info → Run anyway</em>.</li><li><strong>Linux (x86-64, beta):</strong> <code>Kokon_<version>_amd64.deb</code> (Ubuntu, Debian, Mint: <code>sudo apt install ./Kokon_<version>_amd64.deb</code>) or <code>Kokon-<version>-1.x86_64.rpm</code> (Fedora: <code>sudo dnf install ./Kokon-<version>-1.x86_64.rpm</code>). NVIDIA cards need the proprietary driver.</li><li><strong>Mac (Apple Silicon, beta):</strong> <code>Kokon_<version>_aarch64.dmg</code>. Drag Kokon to Applications. The beta is not notarized by Apple yet, so macOS blocks it the first time:</li></ul>@@ -183,7 +185,7 @@<ul><li><strong>RTX 5070 Ti (16 GB):</strong> about 24 s for an edit at 1.5K, about 36 s for a new image at native 2K, about 11 s to upscale a 1-megapixel image 4×.</li><li><strong>RTX 5070 (12 GB):</strong> works with the 12 GB preset, a little slower than the 5070 Ti.</li>- <li><strong>M2 Pro, 16 GB:</strong> about 4 minutes for a 1-megapixel image.</li>+ <li><strong>M2 Pro, 16 GB:</strong> about 65 seconds for a new 1-megapixel image with Kokon's own Mac engine (about 106 s for an edit).</li></ul><p>Kokon measures your machine with the first image and shows realistic time estimates from then on.</p>diff --git a/site/remove-objects-from-photos-offline/index.html b/site/remove-objects-from-photos-offline/index.htmlindex ad24eb3..5bc9ebf 100644--- a/site/remove-objects-from-photos-offline/index.html+++ b/site/remove-objects-from-photos-offline/index.html@@ -22,7 +22,7 @@<link rel="preconnect" href="https://fonts.googleapis.com"><link rel="preconnect" href="https://fonts.gstatic.com" crossorigin><link href="https://fonts.googleapis.com/css2?family=Inter:opsz,[email protected],400..700&family=JetBrains+Mono:wght@400;500&display=swap" rel="stylesheet">-<link rel="stylesheet" href="../assets/style.css?v=13">+<link rel="stylesheet" href="../assets/style.css?v=14"><script type="application/ld+json">{"@context": "https://schema.org",@@ -126,7 +126,8 @@<div class="dd-panel" id="dd-product"><ul><li><a href="/#examples"><b>Examples</b><span>Real edits, before and after</span></a></li><li><a href="/#features"><b>Features</b><span>Edit, paint, combine, expand, upscale</span></a></li>- <li><a href="/#devices"><b>On your phone <span class="new">New</span></b><span>Use your GPU from any device</span></a></li>+ <li><a href="/#layers"><b>Layers lab <span class="new">New</span></b><span>Split an image, move the pieces</span></a></li>+ <li><a href="/#devices"><b>On your phone</b><span>Use your GPU from any device</span></a></li><li><a href="/mcp-image-generation/"><b>For AI agents</b><span>MCP server and HTTP API</span></a></li><li><a href="/#requirements"><b>Requirements</b><span>GPUs, Macs and disk space</span></a></li></ul></div></li>@@ -136,6 +137,7 @@<li><a href="/self-hosted/"><b>Phone and server</b><span>Other devices, Linux, Docker, Tailscale</span></a></li><li><a href="/prompt-guide/"><b>Prompt guide</b><span>Write prompts that work</span></a></li><li><a href="/remove-objects-from-photos-offline/"><b>Remove objects</b><span>Step by step, offline</span></a></li>+ <li><a href="/image-to-layers/"><b>Image to layers <span class="new">New</span></b><span>Split a photo, export a PSD</span></a></li></ul></div></li><li class="dd"><button type="button" aria-expanded="false" aria-controls="dd-compare">Compare<svg class="caret" viewBox="0 0 12 12" aria-hidden="true"><path d="M3 4.5 6 7.5 9 4.5"/></svg></button><div class="dd-panel" id="dd-compare"><ul>@@ -203,7 +205,7 @@</tbody></table></div>- <p>Speed with the default model: on an RTX 5070 Ti, about 24 seconds for an edit at 1.5K and about 36 seconds for a new image at native 2K. 12 GB cards are a little slower, 8 GB cards take 1 to 2 minutes per edit, and an M2 Pro Mac takes about 4 minutes per image.</p>+ <p>Speed with the default model: on an RTX 5070 Ti, about 24 seconds for an edit at 1.5K and about 36 seconds for a new image at native 2K. 12 GB cards are a little slower, 8 GB cards take 1 to 2 minutes per edit, and an M2 Pro Mac takes about a minute for a new 1-megapixel image (about 106 s for an edit) with Kokon's own Mac engine.</p><h2>Questions</h2><div class="faq">@@ -236,6 +238,7 @@<a href="../offline-alternative-to-nano-banana/">Offline Nano Banana alternative</a><a href="../comfyui-alternative/">Simple ComfyUI alternative</a><a href="../remove-objects-from-photos-offline/">Remove objects offline</a>+ <a href="../image-to-layers/">Image to layers</a><a href="../qwen-image-2-1-local/">Run Qwen-Image locally</a><a href="../press/">Press kit</a><a href="../support/">Support Kokon</a>diff --git a/site/self-hosted/index.html b/site/self-hosted/index.htmlindex 1b63487..b6ddc14 100644--- a/site/self-hosted/index.html+++ b/site/self-hosted/index.html@@ -22,7 +22,7 @@<link rel="preconnect" href="https://fonts.googleapis.com"><link rel="preconnect" href="https://fonts.gstatic.com" crossorigin><link href="https://fonts.googleapis.com/css2?family=Inter:opsz,[email protected],400..700&family=JetBrains+Mono:wght@400;500&display=swap" rel="stylesheet">-<link rel="stylesheet" href="../assets/style.css?v=13">+<link rel="stylesheet" href="../assets/style.css?v=14"><link rel="stylesheet" href="../assets/devices.css?v=1"><script type="application/ld+json">{@@ -78,7 +78,8 @@<div class="dd-panel" id="dd-product"><ul><li><a href="/#examples"><b>Examples</b><span>Real edits, before and after</span></a></li><li><a href="/#features"><b>Features</b><span>Edit, paint, combine, expand, upscale</span></a></li>- <li><a href="/#devices"><b>On your phone <span class="new">New</span></b><span>Use your GPU from any device</span></a></li>+ <li><a href="/#layers"><b>Layers lab <span class="new">New</span></b><span>Split an image, move the pieces</span></a></li>+ <li><a href="/#devices"><b>On your phone</b><span>Use your GPU from any device</span></a></li><li><a href="/mcp-image-generation/"><b>For AI agents</b><span>MCP server and HTTP API</span></a></li><li><a href="/#requirements"><b>Requirements</b><span>GPUs, Macs and disk space</span></a></li></ul></div></li>@@ -88,6 +89,7 @@<li><a href="/self-hosted/"><b>Phone and server</b><span>Other devices, Linux, Docker, Tailscale</span></a></li><li><a href="/prompt-guide/"><b>Prompt guide</b><span>Write prompts that work</span></a></li><li><a href="/remove-objects-from-photos-offline/"><b>Remove objects</b><span>Step by step, offline</span></a></li>+ <li><a href="/image-to-layers/"><b>Image to layers <span class="new">New</span></b><span>Split a photo, export a PSD</span></a></li></ul></div></li><li class="dd"><button type="button" aria-expanded="false" aria-controls="dd-compare">Compare<svg class="caret" viewBox="0 0 12 12" aria-hidden="true"><path d="M3 4.5 6 7.5 9 4.5"/></svg></button><div class="dd-panel" id="dd-compare"><ul>diff --git a/site/sitemap.xml b/site/sitemap.xmlindex f7c102a..c9701c8 100644--- a/site/sitemap.xml+++ b/site/sitemap.xml@@ -2,7 +2,7 @@<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9" xmlns:video="http://www.google.com/schemas/sitemap-video/1.1" xmlns:image="http://www.google.com/schemas/sitemap-image/1.1"><url><loc>https://kokon.app/</loc>- <lastmod>2026-10-22</lastmod>+ <lastmod>2026-11-05</lastmod><video:video><video:thumbnail_loc>https://kokon.app/assets/kokon-poster.jpg?v=3</video:thumbnail_loc><video:title>Kokon: edit photos by describing the change, on your own GPU</video:title>@@ -10,6 +10,25 @@<video:content_loc>https://kokon.app/assets/kokon.mp4?v=3</video:content_loc><video:duration>85</video:duration></video:video>+ <video:video>+ <video:thumbnail_loc>https://kokon.app/assets/img/videos/usecases.webp</video:thumbnail_loc>+ <video:title>Free AI photo editor on your PC: remove people, restore old photos, change colors</video:title>+ <video:description>What Kokon can do, in examples made with it: remove photobombers, fix closed eyes, restore and colorize an old photo, add someone who is missing, try a haircut or a jacket, repaint a wall.</video:description>+ <video:content_loc>https://kokon.app/assets/usecases.mp4?v=1</video:content_loc>+ <video:player_loc>https://www.youtube.com/embed/WDbHxkbpU4s</video:player_loc>+ <video:duration>85</video:duration>+ </video:video>+ </url>+ <url>+ <loc>https://kokon.app/image-to-layers/</loc>+ <lastmod>2026-11-05</lastmod>+ <video:video>+ <video:thumbnail_loc>https://kokon.app/assets/img/videos/update-0.12.webp</video:thumbnail_loc>+ <video:title>Kokon 0.12: split any image into layers, move objects, export to PSD</video:title>+ <video:description>The Layers lab: objects cut out onto their own layers, the background filled behind them, layers moved and edited, a layered PSD exported.</video:description>+ <video:player_loc>https://www.youtube.com/embed/eGr9yYe7NVc</video:player_loc>+ <video:duration>141</video:duration>+ </video:video></url><url><loc>https://kokon.app/qwen-image-2-1-local/</loc><lastmod>2026-10-10</lastmod></url><url>diff --git a/site/support/index.html b/site/support/index.htmlindex e95b0a3..f612db4 100644--- a/site/support/index.html+++ b/site/support/index.html@@ -22,7 +22,7 @@<link rel="preconnect" href="https://fonts.googleapis.com"><link rel="preconnect" href="https://fonts.gstatic.com" crossorigin><link href="https://fonts.googleapis.com/css2?family=Inter:opsz,[email protected],400..700&family=JetBrains+Mono:wght@400;500&display=swap" rel="stylesheet">-<link rel="stylesheet" href="../assets/style.css?v=13">+<link rel="stylesheet" href="../assets/style.css?v=14"></head><body><svg width="0" height="0" style="position:absolute" aria-hidden="true">@@ -44,7 +44,8 @@<div class="dd-panel" id="dd-product"><ul><li><a href="/#examples"><b>Examples</b><span>Real edits, before and after</span></a></li><li><a href="/#features"><b>Features</b><span>Edit, paint, combine, expand, upscale</span></a></li>- <li><a href="/#devices"><b>On your phone <span class="new">New</span></b><span>Use your GPU from any device</span></a></li>+ <li><a href="/#layers"><b>Layers lab <span class="new">New</span></b><span>Split an image, move the pieces</span></a></li>+ <li><a href="/#devices"><b>On your phone</b><span>Use your GPU from any device</span></a></li><li><a href="/mcp-image-generation/"><b>For AI agents</b><span>MCP server and HTTP API</span></a></li><li><a href="/#requirements"><b>Requirements</b><span>GPUs, Macs and disk space</span></a></li></ul></div></li>@@ -54,6 +55,7 @@<li><a href="/self-hosted/"><b>Phone and server</b><span>Other devices, Linux, Docker, Tailscale</span></a></li><li><a href="/prompt-guide/"><b>Prompt guide</b><span>Write prompts that work</span></a></li><li><a href="/remove-objects-from-photos-offline/"><b>Remove objects</b><span>Step by step, offline</span></a></li>+ <li><a href="/image-to-layers/"><b>Image to layers <span class="new">New</span></b><span>Split a photo, export a PSD</span></a></li></ul></div></li><li class="dd"><button type="button" aria-expanded="false" aria-controls="dd-compare">Compare<svg class="caret" viewBox="0 0 12 12" aria-hidden="true"><path d="M3 4.5 6 7.5 9 4.5"/></svg></button><div class="dd-panel" id="dd-compare"><ul>@@ -111,6 +113,7 @@<a href="../offline-alternative-to-nano-banana/">Offline Nano Banana alternative</a><a href="../comfyui-alternative/">Simple ComfyUI alternative</a><a href="../remove-objects-from-photos-offline/">Remove objects offline</a>+ <a href="../image-to-layers/">Image to layers</a><a href="../qwen-image-2-1-local/">Run Qwen-Image locally</a><a href="../press/">Press kit</a><a href="../support/">Support Kokon</a>