The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface. The fastest local inference engine in the world.
mm: reset_cast_buffers: sync compute stream before free (#12822)
rattusbcf1a1fab1e9
diff --git a/comfy/model_management.py b/comfy/model_management.pyindex 39b4aa483..07bc8ad67 100644--- a/comfy/model_management.py+++ b/comfy/model_management.py@@ -1148,6 +1148,7 @@ def reset_cast_buffers():LARGEST_CASTED_WEIGHT = (None, 0)for offload_stream in STREAM_CAST_BUFFERS:offload_stream.synchronize()+ synchronize()STREAM_CAST_BUFFERS.clear()soft_empty_cache()