Local UI to run and train LLMs and diffusion models. Supports GGUF, MLX, Qwen3.8, DeepSeek-V4.1, MiniMax-H3, Gemma 4, FLUX and more.
Python agentaichatgptdeepseekfine-tuninggemmaimage-generationllamallmllmsopenaipythonqwenreinforcement-learningself-hostedstable-diffusiontext-to-speechttsuiunsloth
Give the startup transfer budget headroom again after #14524 set it to its own build (#14571)
Daniel Hancbdb0be63fad
diff --git a/studio/frontend/scripts/check-bundle-budget.ts b/studio/frontend/scripts/check-bundle-budget.tsindex 9a311e09a7e..2196f61138c 100644--- a/studio/frontend/scripts/check-bundle-budget.ts+++ b/studio/frontend/scripts/check-bundle-budget.ts@@ -350,7 +350,10 @@ export const BUDGET = {// 2,071,626 B transfer against 4df9ce74f, one machine and build, Node 24, 97 chunks). Raised by that.// The NPU setup card (#14524): +439 B raw / +122 B transfer against fc91656ff (Node 24, 97 chunks), where main// was already 1 B transfer over; transfer set to the measured build, raw raised by the growth.- transferBytes: 2_071_816,+ // #14524 set transfer to its own measured build with no headroom; merged next to #14508 the startup set+ // compresses 7 B larger (2_071_823 B at 5ab80ffdb4, Node 24), so main is over. Measured + 189 B, raw unchanged+ // (already 244 B above the measured 6_879_345).+ transferBytes: 2_072_012,rawBytes: 6_879_589,};