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 data settings driver's cold first load the same 120s as its mount wait (#13809)
Daniel Hanfd8d3bb139e7
diff --git a/tests/studio/playwright_data_settings.py b/tests/studio/playwright_data_settings.pyindex 17a10ca2a22..0f61366e635 100644--- a/tests/studio/playwright_data_settings.py+++ b/tests/studio/playwright_data_settings.py@@ -948,7 +948,11 @@ def main():browser = getattr(p, engine).launch(**options)page = browser.new_page(viewport = {"width": 1280, "height": 1000})page.add_init_script(FIXTURE)- page.goto(url, wait_until = "domcontentloaded")+ # The first load runs Vite's dependency optimizer and transforms the whole module graph+ # from a cold cache, and on Windows CI two of these drivers do that at once. Module+ # scripts run before DOMContentLoaded, so the default 30s has timed out there on a page+ # that then loads fine; allow the same 120s the mount wait below already does.+ page.goto(url, wait_until = "domcontentloaded", timeout = 120_000)page.wait_for_function("!!window.__settingsSmoke", timeout = 120000)output.parent.mkdir(parents = True, exist_ok = True)result = {"engine": engine, "version": browser.version}