Skip to content

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.

Format studio/backend/routes/inference.py back to the hook's fixed point (#13765)

Daniel Han6694ff824607
diff --git a/studio/backend/routes/inference.py b/studio/backend/routes/inference.py
index f576d0d1138..cc132c68009 100644
--- a/studio/backend/routes/inference.py
+++ b/studio/backend/routes/inference.py
@@ -47246,6 +47246,7 @@ async def load_diffusion_model_gated(
)
if pending_name == ENGINE_SD_CPP and vulkan_on_cpu_torch:
needs_gpu = True
+
# A GPU torch whose GPU builds have all failed here runs the managed CPU bundle: no claim.
def _runs_cpu_rung() -> bool:
try: