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
README and site: colored badge icons, one-line hero badge on phones, re-measured times
Gabriele Lozupone94897ead6729
diff --git a/README.md b/README.mdindex 13c7898..63d8137 100644--- a/README.md+++ b/README.md@@ -13,19 +13,19 @@</p><p align="center">- <a href="https://github.com/GabrieleLozupone/kokon/releases/latest"><img src="https://img.shields.io/github/v/release/GabrieleLozupone/kokon?style=flat-square&color=6b6b6b&labelColor=333&label=release" alt="Latest release"></a>- <a href="https://github.com/GabrieleLozupone/kokon/releases"><img src="https://img.shields.io/github/downloads/GabrieleLozupone/kokon/total?style=flat-square&cacheSeconds=3600&color=6b6b6b&labelColor=333" alt="Downloads"></a>- <img src="https://img.shields.io/badge/Windows-10%20%7C%2011-6b6b6b?labelColor=333&style=flat-square&logo=windows&logoColor=white" alt="Windows 10 and 11">- <img src="https://img.shields.io/badge/macOS-26%2B-6b6b6b?labelColor=333&style=flat-square&logo=apple&logoColor=white" alt="macOS 26 or newer">+ <a href="https://github.com/GabrieleLozupone/kokon/releases/latest"><img src="https://img.shields.io/github/v/release/GabrieleLozupone/kokon?style=flat-square&labelColor=333&color=6b6b6b&label=release&cacheSeconds=3600&logo=data%3Aimage%2Fsvg%2Bxml%3Bbase64%2CPHN2ZyB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciIHZpZXdCb3g9IjAgMCAyNCAyNCIgZmlsbD0ibm9uZSIgc3Ryb2tlPSIjQzI5N0ZGIiBzdHJva2Utd2lkdGg9IjIuNCIgc3Ryb2tlLWxpbmVjYXA9InJvdW5kIiBzdHJva2UtbGluZWpvaW49InJvdW5kIj48cGF0aCBkPSJNMTIuNiAyLjZBMiAyIDAgMCAwIDExLjIgMkg0YTIgMiAwIDAgMC0yIDJ2Ny4yYTIgMiAwIDAgMCAuNiAxLjRsOC43IDguN2EyLjQgMi40IDAgMCAwIDMuNCAwbDYuNi02LjZhMi40IDIuNCAwIDAgMCAwLTMuNHoiLz48Y2lyY2xlIGN4PSI3LjUiIGN5PSI3LjUiIHI9IjEuMiIgZmlsbD0iI0MyOTdGRiIvPjwvc3ZnPg%3D%3D" alt="Latest release"></a>+ <a href="https://github.com/GabrieleLozupone/kokon/releases"><img src="https://img.shields.io/github/downloads/GabrieleLozupone/kokon/total?style=flat-square&labelColor=333&color=6b6b6b&cacheSeconds=3600&logo=data%3Aimage%2Fsvg%2Bxml%3Bbase64%2CPHN2ZyB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciIHZpZXdCb3g9IjAgMCAyNCAyNCIgZmlsbD0ibm9uZSIgc3Ryb2tlPSIjM0ZCOTUwIiBzdHJva2Utd2lkdGg9IjIuNCIgc3Ryb2tlLWxpbmVjYXA9InJvdW5kIiBzdHJva2UtbGluZWpvaW49InJvdW5kIj48cGF0aCBkPSJNMTIgM3YxMiIvPjxwYXRoIGQ9Im03IDEwIDUgNSA1LTUiLz48cGF0aCBkPSJNNCAyMWgxNiIvPjwvc3ZnPg%3D%3D" alt="Downloads"></a>+ <img src="https://img.shields.io/badge/Windows-10%20%7C%2011-6b6b6b?style=flat-square&labelColor=333&color=6b6b6b&logo=data%3Aimage%2Fsvg%2Bxml%3Bbase64%2CPHN2ZyB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciIHZpZXdCb3g9IjAgMCAyNCAyNCIgZmlsbD0iIzAwQTRFRiI%2BPHBhdGggZD0iTTIgMmg5LjV2OS41SDJ6TTEyLjUgMkgyMnY5LjVoLTkuNXpNMiAxMi41aDkuNVYyMkgyek0xMi41IDEyLjVIMjJWMjJoLTkuNXoiLz48L3N2Zz4%3D" alt="Windows 10 and 11">+ <img src="https://img.shields.io/badge/macOS-26%2B-6b6b6b?style=flat-square&labelColor=333&color=6b6b6b&logo=apple&logoColor=white" alt="macOS 26 or newer"><br>- <img src="https://img.shields.io/badge/NVIDIA-12%20GB%2B%20VRAM-6b6b6b?labelColor=333&style=flat-square&logo=nvidia&logoColor=76B900" alt="NVIDIA, 12 GB of VRAM or more">- <img src="https://img.shields.io/badge/AMD-12%20GB%2B%20VRAM%20(beta)-6b6b6b?labelColor=333&style=flat-square&logo=amd&logoColor=white" alt="AMD, 12 GB of VRAM or more, beta">- <img src="https://img.shields.io/badge/Intel%20Arc-12%20GB%2B%20VRAM%20(beta)-6b6b6b?labelColor=333&style=flat-square&logo=intel&logoColor=white" alt="Intel Arc, 12 GB of VRAM or more, beta">- <img src="https://img.shields.io/badge/Apple%20Silicon-32%20GB%20recommended%20(beta)-6b6b6b?labelColor=333&style=flat-square&logo=apple&logoColor=white" alt="Apple Silicon, 32 GB of unified memory recommended, beta">+ <img src="https://img.shields.io/badge/NVIDIA-12%20GB%2B%20VRAM-6b6b6b?style=flat-square&labelColor=333&color=6b6b6b&logo=nvidia&logoColor=76B900" alt="NVIDIA, 12 GB of VRAM or more">+ <img src="https://img.shields.io/badge/AMD-12%20GB%2B%20VRAM%20(beta)-6b6b6b?style=flat-square&labelColor=333&color=6b6b6b&logo=amd&logoColor=ED1C24" alt="AMD, 12 GB of VRAM or more, beta">+ <img src="https://img.shields.io/badge/Intel%20Arc-12%20GB%2B%20VRAM%20(beta)-6b6b6b?style=flat-square&labelColor=333&color=6b6b6b&logo=intel&logoColor=00C7FD" alt="Intel Arc, 12 GB of VRAM or more, beta">+ <img src="https://img.shields.io/badge/Apple%20Silicon-32%20GB%20recommended%20(beta)-6b6b6b?style=flat-square&labelColor=333&color=6b6b6b&logo=apple&logoColor=white" alt="Apple Silicon, 32 GB of unified memory recommended, beta"><br>- <a href="#use-it-from-your-ai-agent"><img src="https://img.shields.io/badge/MCP-server%20included-6b6b6b?labelColor=333&style=flat-square" alt="MCP server included"></a>- <a href="LICENSE"><img src="https://img.shields.io/github/license/GabrieleLozupone/kokon?style=flat-square&cacheSeconds=3600&color=6b6b6b&labelColor=333" alt="License"></a>- <a href="https://github.com/GabrieleLozupone/kokon/stargazers"><img src="https://img.shields.io/github/stars/GabrieleLozupone/kokon?style=flat-square&cacheSeconds=3600&color=6b6b6b&labelColor=333" alt="Stars"></a>+ <a href="#use-it-from-your-ai-agent"><img src="https://img.shields.io/badge/MCP-server%20included-6b6b6b?style=flat-square&labelColor=333&color=6b6b6b&logo=modelcontextprotocol&logoColor=white" alt="MCP server included"></a>+ <a href="LICENSE"><img src="https://img.shields.io/badge/license-AGPL--3.0-6b6b6b?style=flat-square&labelColor=333&color=6b6b6b&logo=data%3Aimage%2Fsvg%2Bxml%3Bbase64%2CPHN2ZyB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciIHZpZXdCb3g9IjAgMCAyNCAyNCIgZmlsbD0ibm9uZSIgc3Ryb2tlPSIjNThBNkZGIiBzdHJva2Utd2lkdGg9IjIuNCIgc3Ryb2tlLWxpbmVjYXA9InJvdW5kIiBzdHJva2UtbGluZWpvaW49InJvdW5kIj48cGF0aCBkPSJNMTIgM3YxOCIvPjxwYXRoIGQ9Ik03IDIxaDEwIi8%2BPHBhdGggZD0iTTMgN2gxOCIvPjxwYXRoIGQ9Im02IDctMyA3YTMuNSAzLjUgMCAwIDAgNiAweiIvPjxwYXRoIGQ9Im0xOCA3LTMgN2EzLjUgMy41IDAgMCAwIDYgMHoiLz48L3N2Zz4%3D" alt="License: AGPL-3.0"></a>+ <a href="https://github.com/GabrieleLozupone/kokon/stargazers"><img src="https://img.shields.io/github/stars/GabrieleLozupone/kokon?style=flat-square&labelColor=333&color=6b6b6b&cacheSeconds=3600&logo=data%3Aimage%2Fsvg%2Bxml%3Bbase64%2CPHN2ZyB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciIHZpZXdCb3g9IjAgMCAyNCAyNCIgZmlsbD0ibm9uZSIgc3Ryb2tlPSIjRjVDNTE4IiBzdHJva2Utd2lkdGg9IjIuNCIgc3Ryb2tlLWxpbmVjYXA9InJvdW5kIiBzdHJva2UtbGluZWpvaW49InJvdW5kIj48cGF0aCBmaWxsPSIjRjVDNTE4IiBkPSJNMTIgMi41bDIuOSA2IDYuNi45LTQuOCA0LjYgMS4yIDYuNUwxMiAxNy40bC01LjkgMy4xIDEuMi02LjVMMi41IDkuNGw2LjYtLjl6Ii8%2BPC9zdmc%2B" alt="Stars"></a></p><p align="center">@@ -148,7 +148,7 @@ whole thing takes a few minutes, and the setup screen shows what Kokon can do wh| **GPU** | NVIDIA, 12 GB VRAM (RTX 3060 12 GB, 4070) · AMD / Intel 12 GB+ through Vulkan (beta) | NVIDIA, 16 GB+ VRAM (RTX 4080, 5070 Ti, 5080, 4090, 5090) || **RAM** | 16 GB | 32 GB || **Disk** | 15 GB free | 25 GB free on an SSD |-| **Time per edit** <!-- TODO re-measure at CFG 1 before launch; RTX 5070 12 GB not timed yet --> | A little slower than the RTX 5070 Ti on a 12 GB RTX 5070 | About 1 minute at Balanced on a 16 GB RTX 5070 Ti |+| **Time per image** <!-- RTX 5070 edit time not measured yet (text to image only, 10-11) --> | RTX 5070 12 GB: ~12 s per 1 MP image, ~42 s at native 2K (text to image) | RTX 5070 Ti 16 GB: ~24 s per edit at Balanced (1.5K), ~36 s for a new image at native 2K |**Mac (beta)**diff --git a/docs/engineering-notes.md b/docs/engineering-notes.mdindex 83a22ab..764a072 100644--- a/docs/engineering-notes.md+++ b/docs/engineering-notes.md@@ -192,3 +192,14 @@ edit, inpaint, outpaint, multi-reference and upscale through an isolated servicetime, working size, peak VRAM, memory spilled to shared GPU memory and out-of-memory retriesinto `docs\test-results\`. Bandwidth and compute stay those of the bigger card, so times areoptimistic: a real 12 GB card (RTX 3060/4070) has to confirm them.++Real RTX 5070 12 GB (a friend's diagnostics, 2026-10-11; 12gb preset, Turbo Q4_K, text encoder on+the CPU, text to image only): 1376×768 at 1.24 s/step (10 s sampling, 1.9 s decode); ~3.2 MP+(2400×1344, 1216×2624) at 4.3 s/step (35 s sampling, 7.2 s tiled decode). No OOM retries and no+prefix-cache fallback; ~5.8 GB VRAM free with the model loaded. About 1.5× the 5070 Ti's step time,+so `planner::step_seconds` (calibrated on the Ti) underestimates on this card.++RTX 5070 Ti 16 GB (app diagnostics, 2026-10-11; 16gb preset, Turbo Q4_K, text encoder on the CPU):+text to image 2048² at 3.67 s/step (29.8 s sampling, 5.9 s tiled decode), 36 s total; edit at+1472² with one reference at 2.0 s/step (1.0 s encode, 1.7 s conditioning, 18.3 s sampling, 2.6 s+decode), 24 s total; ESRGAN 4× of 1448² to 5792² in 22 s.diff --git a/site/assets/style.css b/site/assets/style.cssindex 8e4c1e9..f6e95da 100644--- a/site/assets/style.css+++ b/site/assets/style.css@@ -55,6 +55,8 @@ body::before { content: ""; position: fixed; inset: 0; pointer-events: none; z-i.hero::before { content: ""; position: absolute; left: 50%; top: -240px; width: 1400px; height: 900px; transform: translateX(-50%); background: radial-gradient(ellipse at center, rgba(247,246,242,0.09) 0%, rgba(247,246,242,0) 60%); pointer-events: none; }.eyebrow { display: inline-flex; align-items: center; gap: 10px; padding: 6px 14px 6px 6px; border-radius: 999px; background: var(--card); box-shadow: inset 0 0 0 1px var(--line); font-size: 13.5px; color: var(--text-2); }.eyebrow b { font-weight: 600; color: var(--ink); background: var(--paper); padding: 3px 9px; border-radius: 999px; font-size: 12px; }+.eyebrow > span { white-space: nowrap; }+@media (max-width: 479px) { .eyebrow .eb-long { display: none; } }h1 { font-size: clamp(44px, 8.4vw, 104px); line-height: 1; letter-spacing: -0.045em; font-weight: 680; margin: 26px auto 0; max-width: 11.5em; }h1 .soft { color: var(--text-3); }.lede { font-size: clamp(17px, 2vw, 21px); color: var(--text-2); max-width: 36em; margin: 26px auto 0; letter-spacing: -0.012em; }diff --git a/site/index.html b/site/index.htmlindex eeeb9a7..588a1bc 100644--- a/site/index.html+++ b/site/index.html@@ -222,7 +222,7 @@<main id="top"><section class="hero"><div class="wrap">- <span class="eyebrow rv"><b>Beta</b>Qwen-Image 2.1 Turbo, on your own computer</span>+ <span class="eyebrow rv"><b>Beta</b><span>Qwen-Image 2.1 Turbo<span class="eb-long">, on your own computer</span></span></span><h1 class="rv d1">Edit photos by describing the change.</h1><p class="lede rv d2">Kokon runs Qwen-Image 2.1, the best open image-editing model, on your own GPU. One installer for Windows and Mac. No Python, no ComfyUI, no cloud, no account.</p><div class="ctas rv d3" id="download">@@ -276,7 +276,7 @@<span class="send" aria-hidden="true"><svg width="18" height="18" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2.4" stroke-linecap="round" stroke-linejoin="round"><path d="M12 19V5M5 12l7-7 7 7"/></svg></span></div></div>- <p class="honest">Times on an RTX 5070 Ti: 22 to 46 s for a new image, 25 to 55 s for an edit. The prompts and seeds are in the repository.</p>+ <p class="honest">Times on an RTX 5070 Ti: about 36 s for a new image at native 2K, about 24 s for an edit at 1.5K. The prompts and seeds are in the repository.</p></div></section>diff --git a/site/mcp-image-generation/index.html b/site/mcp-image-generation/index.htmlindex 8ee67ee..0df5dd7 100644--- a/site/mcp-image-generation/index.html+++ b/site/mcp-image-generation/index.html@@ -148,7 +148,7 @@</ul><h2>Requirements</h2>- <p>A Windows PC with a 12 GB NVIDIA, AMD or Intel Arc GPU, or an Apple Silicon Mac (32 GB recommended). On an RTX 5070 Ti a new image takes 22 to 46 s and an edit 25 to 55 s. Details in the <a href="../qwen-image-2-1-local/">setup guide</a>.</p>+ <p>A Windows PC with a 12 GB NVIDIA, AMD or Intel Arc GPU, or an Apple Silicon Mac (32 GB recommended). On an RTX 5070 Ti a new image at native 2K takes about 36 s and an edit at 1.5K about 24 s. Details in the <a href="../qwen-image-2-1-local/">setup guide</a>.</p><div class="cta-box"><h2>Give your agent a GPU</h2>diff --git a/site/qwen-image-2-1-local/index.html b/site/qwen-image-2-1-local/index.htmlindex 4500fde..b1f3e8f 100644--- a/site/qwen-image-2-1-local/index.html+++ b/site/qwen-image-2-1-local/index.html@@ -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). 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 22 s for a 1.5K image, about 40 s for native 2K, and 25 to 55 s for an edit. 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 4 minutes for a 1-megapixel image." } }]}]@@ -147,7 +147,7 @@<h2>How fast is it?</h2><p>Measured with Qwen-Image 2.1 Turbo (8 steps):</p><ul>- <li><strong>RTX 5070 Ti (16 GB):</strong> about 22 s for a new 1.5K image, about 40 s at native 2K, 25 to 55 s for an edit.</li>+ <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 22 s for a 4× upscale.</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></ul>