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
MLX inference stub: answer the _FIRST_USE_WORK import zoo added (#16537)
Daniel Han867c4248d0ae
diff --git a/studio/backend/tests/test_mlx_inference_backend.py b/studio/backend/tests/test_mlx_inference_backend.pyindex 2484deb6cb3..103efc5246b 100644--- a/studio/backend/tests/test_mlx_inference_backend.py+++ b/studio/backend/tests/test_mlx_inference_backend.py@@ -132,11 +132,12 @@ def _fusion_modules(model, modules):# Plain-function helpers zoo's generate.py imports; the neutral scope cannot stand in for them.PLAIN_HELPERS = {"_fusion_modules": _fusion_modules,- # Private names zoo's speculative.py imports (unsloth_zoo #2330); only their presence matters here.+ # Private names zoo's speculative.py imports (unsloth_zoo #2330, #2365); only their presence matters here."_SparseDrafter": type("_SparseDrafter", (), {"__init__": lambda self, *_args, **_kwargs: None}),"_VERIFY_GDN_MAX_ROWS": 16,+ "_FIRST_USE_WORK": [0],}