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
Advertise NVFP4 diffusion on the text encoder driver's mocked host (#12047)
Daniel Han8faa867ff739
diff --git a/tests/studio/playwright_image_text_encoder.py b/tests/studio/playwright_image_text_encoder.pyindex 30a9e93ce30..75b3464efa5 100644--- a/tests/studio/playwright_image_text_encoder.py+++ b/tests/studio/playwright_image_text_encoder.py@@ -175,6 +175,13 @@ def main():"error": None,},)+ elif path == "/api/system":+ # NVFP4 options only render on a host that reports NVFP4 diffusion (the options go+ # through withNvfp4Option since 8ecfd3092), and the loop below picks NVFP4, so this+ # host is a Blackwell one.+ system = _api_payload(path, query, full_footprint = True)+ system["nvfp4_diffusion"] = True+ _json(route, system)else:_json(route, _api_payload(path, query, full_footprint = True))