Skip to content

armgit / kokon

Public Imported from GitHub
0

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.

ui: LoRA marked beta in Settings (tested on Windows only)

Gabriele Lozupone6387ba193c9d
diff --git a/src/components/LoraSettings.tsx b/src/components/LoraSettings.tsx
index e2caa2d..d608f6d 100644
--- a/src/components/LoraSettings.tsx
+++ b/src/components/LoraSettings.tsx
@@ -120,7 +120,7 @@ export function LoraSettings() {
<>
<div className="setting-card">
<div className="setting-card-text">
- <strong>Your LoRA files</strong>
+ <strong>Your LoRA files (beta)</strong>
<span className="muted">
Load LoRAs made for Qwen-Image 2.1 (.safetensors). Kokon checks each file and keeps its own prepared copy; your file is not changed. Turn them on or off for each project with the
LoRA button next to the prompt, up to {MAX_LORAS} at a time.
@@ -130,6 +130,9 @@ export function LoraSettings() {
<Plus size={15} /> {busy ? "Checking…" : "Load a LoRA…"}
</button>
</div>
+ <span className="muted">
+ Beta: tested on Windows with an NVIDIA card. It should work the same on Mac and Linux, but it hasn't been tested there yet: please tell us how it goes on GitHub.
+ </span>
<span className="muted">
Most LoRAs are trained on the full Qwen-Image 2.1 model, while Kokon runs the faster Turbo model. The popular ones we tested didn't improve on what Turbo does with a good prompt, so compare
with the LoRA on and off. A LoRA makes each image about 20 % slower.
diff --git a/src/components/Settings.tsx b/src/components/Settings.tsx
index 0890c73..1784e5e 100644
--- a/src/components/Settings.tsx
+++ b/src/components/Settings.tsx
@@ -232,7 +232,7 @@ export function Settings() {
const nav: [Tab, string, React.ReactNode][] = [
["general", "General", <Settings2 size={16} key="g" />],
["models", "Models", <HardDrive size={16} key="m" />],
- ["lora", "LoRA", <Layers size={16} key="l" />],
+ ["lora", "LoRA (beta)", <Layers size={16} key="l" />],
["performance", "Performance", <Cpu size={16} key="p" />],
["diagnostics", "Diagnostics", <Activity size={16} key="d" />],
];