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
Studio tests: queue each FIFO admission waiter before starting the next (#15457)
Daniel Han92287a4a965f
diff --git a/studio/backend/tests/test_anthropic_admission.py b/studio/backend/tests/test_anthropic_admission.pyindex 898e7e678e1..d06324263dc 100644--- a/studio/backend/tests/test_anthropic_admission.py+++ b/studio/backend/tests/test_anthropic_admission.py@@ -589,8 +589,13 @@ def test_fifo_fairness_across_many_waiters(monkeypatch):order.append(i)return resp- tasks = [asyncio.create_task(_one(i)) for i in range(8)]- await asyncio.sleep(0.2)+ # Admission awaits prompt preparation first: queue each waiter before the next so arrival order is fixed.+ tasks = []+ for i in range(8):+ tasks.append(asyncio.create_task(_one(i)))+ deadline = time.monotonic() + 5+ while _snapshot().queued < i + 1 and time.monotonic() < deadline:+ await asyncio.sleep(0.01)assert _snapshot().queued == 8held[0].release()await asyncio.wait_for(asyncio.gather(*tasks), timeout = 5)