• Joined on 2024-08-30
wangbo synced commits to refs/pull/11000/head at wangbo/ComfyUI from mirror 2025-12-05 12:27:05 +08:00
b8afb60ee8 torch compile fixes.
295a0170d6 Remove.
55366d9c0f Remove.
c217243b56 Fix fp8 fast lora offload.
Compare 4 commits »
wangbo synced commits to refs/pull/10845/merge at wangbo/ComfyUI from mirror 2025-12-05 12:27:05 +08:00
3c8456223c [API Nodes]: fixes and refactor (#11104)
Compare 2 commits »
wangbo synced commits to refs/pull/10258/merge at wangbo/ComfyUI from mirror 2025-12-05 12:27:05 +08:00
3c8456223c [API Nodes]: fixes and refactor (#11104)
Compare 2 commits »
wangbo synced and deleted reference refs/tags/refs/pull/11104/merge at wangbo/ComfyUI from mirror 2025-12-05 12:27:04 +08:00
wangbo synced commits to master at wangbo/ComfyUI from mirror 2025-12-05 12:27:04 +08:00
35fa091340 Forgot to put this in README. (#11112)
3c8456223c [API Nodes]: fixes and refactor (#11104)
Compare 2 commits »
wangbo synced commits to fp8_refactor at wangbo/ComfyUI from mirror 2025-12-05 12:27:04 +08:00
b8afb60ee8 torch compile fixes.
295a0170d6 Remove.
55366d9c0f Remove.
c217243b56 Fix fp8 fast lora offload.
Compare 4 commits »
wangbo synced commits to refs/pull/9305/merge at wangbo/ComfyUI from mirror 2025-12-05 04:17:08 +08:00
8ef6aeaceb Merge 32d2ece3fe1c52521f3533f4743b782c3e2bb284 into 9bc893c5bb
9bc893c5bb sd: bump HY1.5 VAE estimate (#11107)
f4bdf5f830 sd: revise hy VAE VRAM (#11105)
6be85c7920 mp: use look-ahead actuals for stream offload VRAM calculation (#11096)
ea17add3c6 Fix case where text encoders where running on the CPU instead of GPU. (#11095)
Compare 6 commits »
wangbo synced commits to refs/pull/8289/merge at wangbo/ComfyUI from mirror 2025-12-05 04:17:08 +08:00
9bc893c5bb sd: bump HY1.5 VAE estimate (#11107)
f4bdf5f830 sd: revise hy VAE VRAM (#11105)
Compare 3 commits »
wangbo synced commits to refs/pull/8219/merge at wangbo/ComfyUI from mirror 2025-12-05 04:17:08 +08:00
6be85c7920 mp: use look-ahead actuals for stream offload VRAM calculation (#11096)
ea17add3c6 Fix case where text encoders where running on the CPU instead of GPU. (#11095)
ecdc8697d5 Qwen Image Lora training fix from #11090 (#11094)
dce518c2b4 convert nodes_audio.py to V3 schema (#10798)
Compare 6 commits »
wangbo synced commits to refs/pull/8950/merge at wangbo/ComfyUI from mirror 2025-12-05 04:17:08 +08:00
9bc893c5bb sd: bump HY1.5 VAE estimate (#11107)
f4bdf5f830 sd: revise hy VAE VRAM (#11105)
6be85c7920 mp: use look-ahead actuals for stream offload VRAM calculation (#11096)
ea17add3c6 Fix case where text encoders where running on the CPU instead of GPU. (#11095)
Compare 10 commits »
wangbo synced commits to refs/pull/10986/merge at wangbo/ComfyUI from mirror 2025-12-05 04:17:07 +08:00
9bc893c5bb sd: bump HY1.5 VAE estimate (#11107)
f4bdf5f830 sd: revise hy VAE VRAM (#11105)
6be85c7920 mp: use look-ahead actuals for stream offload VRAM calculation (#11096)
ea17add3c6 Fix case where text encoders where running on the CPU instead of GPU. (#11095)
Compare 7 commits »
wangbo synced commits to refs/pull/10947/merge at wangbo/ComfyUI from mirror 2025-12-05 04:17:07 +08:00
7d3a40270a Merge 12a516c58053f3541bde85a261ece0efd5019720 into 9bc893c5bb
9bc893c5bb sd: bump HY1.5 VAE estimate (#11107)
f4bdf5f830 sd: revise hy VAE VRAM (#11105)
12a516c580 feat(api-nodes): add support for seedance-1-0-pro-fast model
Compare 4 commits »
wangbo synced commits to refs/pull/10904/merge at wangbo/ComfyUI from mirror 2025-12-05 04:17:07 +08:00
9bc893c5bb sd: bump HY1.5 VAE estimate (#11107)
f4bdf5f830 sd: revise hy VAE VRAM (#11105)
Compare 3 commits »
wangbo synced commits to refs/pull/10928/merge at wangbo/ComfyUI from mirror 2025-12-05 04:17:07 +08:00
9bc893c5bb sd: bump HY1.5 VAE estimate (#11107)
f4bdf5f830 sd: revise hy VAE VRAM (#11105)
Compare 3 commits »
wangbo synced commits to refs/pull/10941/merge at wangbo/ComfyUI from mirror 2025-12-05 04:17:07 +08:00
9bc893c5bb sd: bump HY1.5 VAE estimate (#11107)
f4bdf5f830 sd: revise hy VAE VRAM (#11105)
6be85c7920 mp: use look-ahead actuals for stream offload VRAM calculation (#11096)
ea17add3c6 Fix case where text encoders where running on the CPU instead of GPU. (#11095)
Compare 8 commits »
wangbo synced commits to refs/pull/10944/merge at wangbo/ComfyUI from mirror 2025-12-05 04:17:07 +08:00
6490fafc99 Merge 1e6313b15def4ff0e8f2c0d5ff362754e09665e5 into 9bc893c5bb
9bc893c5bb sd: bump HY1.5 VAE estimate (#11107)
f4bdf5f830 sd: revise hy VAE VRAM (#11105)
Compare 3 commits »
wangbo synced commits to refs/pull/10947/head at wangbo/ComfyUI from mirror 2025-12-05 04:17:07 +08:00
12a516c580 feat(api-nodes): add support for seedance-1-0-pro-fast model
6be85c7920 mp: use look-ahead actuals for stream offload VRAM calculation (#11096)
ea17add3c6 Fix case where text encoders where running on the CPU instead of GPU. (#11095)
ecdc8697d5 Qwen Image Lora training fix from #11090 (#11094)
dce518c2b4 convert nodes_audio.py to V3 schema (#10798)
Compare 42 commits »
wangbo synced commits to refs/pull/11026/merge at wangbo/ComfyUI from mirror 2025-12-05 04:17:07 +08:00
9bc893c5bb sd: bump HY1.5 VAE estimate (#11107)
f4bdf5f830 sd: revise hy VAE VRAM (#11105)
Compare 3 commits »
wangbo synced commits to refs/pull/10975/merge at wangbo/ComfyUI from mirror 2025-12-05 04:17:07 +08:00
abedb3cd0c Merge 96eaf4ebcf612df48fc56072dec1e1ee97cb50c6 into 9bc893c5bb
9bc893c5bb sd: bump HY1.5 VAE estimate (#11107)
f4bdf5f830 sd: revise hy VAE VRAM (#11105)
Compare 3 commits »
wangbo synced commits to refs/pull/10988/merge at wangbo/ComfyUI from mirror 2025-12-05 04:17:07 +08:00
9bc893c5bb sd: bump HY1.5 VAE estimate (#11107)
f4bdf5f830 sd: revise hy VAE VRAM (#11105)
Compare 3 commits »