wangbo
0 Followers · 0 Following
  • Joined on 2024-08-30
Repositories
5
 Projects Packages Public Activity Starred Repositories
wangbo synced commits to refs/pull/1038/merge at wangbo/ComfyUI-Manager from mirror 2026-03-26 19:33:38 +08:00
bc87e8feb1 Merge caac2ef276 into 2007eab26d
2007eab26d fix(security): add litellm supply chain attack detection and improve pip matching (#2731)
d870273d4b update DB
Compare 3 commits »
wangbo synced commits to refs/pull/9305/merge at wangbo/ComfyUI from mirror 2026-03-26 11:34:40 +08:00
f35209adee Merge 267e576007 into 6580a6bc01
6580a6bc01 fix(number-convert): preserve int precision for large numbers (#13147)
3b15651bc6 bump manager version to 4.1 (#13156)
a55835f10c fix(api-nodes): made Reve node price badges more precise (#13154)
Compare 4 commits »
wangbo synced commits to refs/pull/9113/merge at wangbo/ComfyUI from mirror 2026-03-26 11:34:40 +08:00
c9bc4f584e Merge 5f74786fd5 into a55835f10c
5f74786fd5 Merge branch 'master' into models-dir-arg
a55835f10c fix(api-nodes): made Reve node price badges more precise (#13154)
b53b10ea61 Fix Train LoRA crash when training_dtype is "none" with bfloat16 LoRA weights (#13145)
7d5534d8e5 feat(assets): register output files as assets after prompt execution (#12812)
Compare 6 commits »
wangbo synced commits to refs/pull/9756/merge at wangbo/ComfyUI from mirror 2026-03-26 11:34:40 +08:00
c5f9b423d7 Merge 4916544c8d into a55835f10c
a55835f10c fix(api-nodes): made Reve node price badges more precise (#13154)
b53b10ea61 Fix Train LoRA crash when training_dtype is "none" with bfloat16 LoRA weights (#13145)
7d5534d8e5 feat(assets): register output files as assets after prompt execution (#12812)
5ebb0c2e0b FP8 bwd training (#13121)
Compare 5 commits »
wangbo synced commits to refs/pull/9113/head at wangbo/ComfyUI from mirror 2026-03-26 11:34:39 +08:00
5f74786fd5 Merge branch 'master' into models-dir-arg
a55835f10c fix(api-nodes): made Reve node price badges more precise (#13154)
b53b10ea61 Fix Train LoRA crash when training_dtype is "none" with bfloat16 LoRA weights (#13145)
7d5534d8e5 feat(assets): register output files as assets after prompt execution (#12812)
5ebb0c2e0b FP8 bwd training (#13121)
Compare 564 commits »
wangbo synced commits to refs/pull/8380/head at wangbo/ComfyUI from mirror 2026-03-26 11:34:38 +08:00
fdbcf4c02b Merge branch 'master' into date-sorted-saving
a40d162b6c resolve merge conflict in nodes_images.py
a55835f10c fix(api-nodes): made Reve node price badges more precise (#13154)
b53b10ea61 Fix Train LoRA crash when training_dtype is "none" with bfloat16 LoRA weights (#13145)
7d5534d8e5 feat(assets): register output files as assets after prompt execution (#12812)
Compare 737 commits »
wangbo synced commits to refs/pull/8380/merge at wangbo/ComfyUI from mirror 2026-03-26 11:34:38 +08:00
08f8148889 Merge fdbcf4c02b into a55835f10c
fdbcf4c02b Merge branch 'master' into date-sorted-saving
a40d162b6c resolve merge conflict in nodes_images.py
a55835f10c fix(api-nodes): made Reve node price badges more precise (#13154)
b53b10ea61 Fix Train LoRA crash when training_dtype is "none" with bfloat16 LoRA weights (#13145)
Compare 571 commits »
wangbo synced commits to refs/pull/13105/merge at wangbo/ComfyUI from mirror 2026-03-26 11:34:37 +08:00
7b2602ccd8 Merge d61ec431c4 into a55835f10c
a55835f10c fix(api-nodes): made Reve node price badges more precise (#13154)
b53b10ea61 Fix Train LoRA crash when training_dtype is "none" with bfloat16 LoRA weights (#13145)
7d5534d8e5 feat(assets): register output files as assets after prompt execution (#12812)
5ebb0c2e0b FP8 bwd training (#13121)
Compare 9 commits »
wangbo synced commits to refs/pull/13146/merge at wangbo/ComfyUI from mirror 2026-03-26 11:34:37 +08:00
0bd3589e4e Merge 40a25de926 into a55835f10c
a55835f10c fix(api-nodes): made Reve node price badges more precise (#13154)
b53b10ea61 Fix Train LoRA crash when training_dtype is "none" with bfloat16 LoRA weights (#13145)
7d5534d8e5 feat(assets): register output files as assets after prompt execution (#12812)
Compare 4 commits »
wangbo synced commits to refs/pull/13153/merge at wangbo/ComfyUI from mirror 2026-03-26 11:34:37 +08:00
6777fb48fa Merge bea22ba97ac98cc6920495c10c951c27ada8fcf8 into 404d7b9978
404d7b9978 feat: Support Qwen3.5 text generation models (#12771)
6580a6bc01 fix(number-convert): preserve int precision for large numbers (#13147)
3b15651bc6 bump manager version to 4.1 (#13156)
a55835f10c fix(api-nodes): made Reve node price badges more precise (#13154)
Compare 5 commits »
wangbo synced commits to refs/pull/7562/merge at wangbo/ComfyUI from mirror 2026-03-26 11:34:37 +08:00
c2268c39eb Merge 6fd6ffd023 into a55835f10c
a55835f10c fix(api-nodes): made Reve node price badges more precise (#13154)
b53b10ea61 Fix Train LoRA crash when training_dtype is "none" with bfloat16 LoRA weights (#13145)
7d5534d8e5 feat(assets): register output files as assets after prompt execution (#12812)
5ebb0c2e0b FP8 bwd training (#13121)
Compare 5 commits »
wangbo synced commits to refs/pull/13135/merge at wangbo/ComfyUI from mirror 2026-03-26 11:34:37 +08:00
b3c4c26b7e Merge 3a82bd15e7 into a55835f10c
a55835f10c fix(api-nodes): made Reve node price badges more precise (#13154)
Compare 2 commits »
wangbo synced commits to refs/pull/13082/head at wangbo/ComfyUI from mirror 2026-03-26 11:34:36 +08:00
c4ce864b9f Add better error handling for a custom ar_video sampler.
e9cf4659d2 Base frame_seq_len on the padded token grid.
08bf8f4d95 Move KV cache end counter to Python int to avoid per-step host synchronization in AR sampling loops.
3440c57f67 Remove ar_convert, now present in hg repackaged model repo.
de66e64ec2 Fix 'Process the tail block instead of truncating it', fix 'Don't mutate the patcher's shared transformer_options in place'.
Compare 34 commits »
wangbo synced commits to refs/pull/13081/merge at wangbo/ComfyUI from mirror 2026-03-26 11:34:36 +08:00
bfa0c616ff Merge d898f2aa0e into a55835f10c
a55835f10c fix(api-nodes): made Reve node price badges more precise (#13154)
b53b10ea61 Fix Train LoRA crash when training_dtype is "none" with bfloat16 LoRA weights (#13145)
7d5534d8e5 feat(assets): register output files as assets after prompt execution (#12812)
5ebb0c2e0b FP8 bwd training (#13121)
Compare 10 commits »
wangbo synced commits to refs/pull/13096/merge at wangbo/ComfyUI from mirror 2026-03-26 11:34:36 +08:00
60061fe1e7 Merge b356974a50 into a55835f10c
a55835f10c fix(api-nodes): made Reve node price badges more precise (#13154)
b53b10ea61 Fix Train LoRA crash when training_dtype is "none" with bfloat16 LoRA weights (#13145)
7d5534d8e5 feat(assets): register output files as assets after prompt execution (#12812)
5ebb0c2e0b FP8 bwd training (#13121)
Compare 9 commits »
wangbo synced commits to refs/pull/13082/merge at wangbo/ComfyUI from mirror 2026-03-26 11:34:36 +08:00
50f5547668 Merge c4ce864b9f1c30e8cb1342ac38a0a41a58920c26 into 3b15651bc6
c4ce864b9f Add better error handling for a custom ar_video sampler.
e9cf4659d2 Base frame_seq_len on the padded token grid.
08bf8f4d95 Move KV cache end counter to Python int to avoid per-step host synchronization in AR sampling loops.
3b15651bc6 bump manager version to 4.1 (#13156)
Compare 14 commits »
wangbo synced commits to refs/pull/13094/merge at wangbo/ComfyUI from mirror 2026-03-26 11:34:36 +08:00
b22724595b Merge 3f627561f0 into a55835f10c
a55835f10c fix(api-nodes): made Reve node price badges more precise (#13154)
b53b10ea61 Fix Train LoRA crash when training_dtype is "none" with bfloat16 LoRA weights (#13145)
7d5534d8e5 feat(assets): register output files as assets after prompt execution (#12812)
Compare 4 commits »
wangbo synced commits to refs/pull/13080/merge at wangbo/ComfyUI from mirror 2026-03-26 11:34:36 +08:00
068c953635 Merge 1f8a74b76f into a55835f10c
a55835f10c fix(api-nodes): made Reve node price badges more precise (#13154)
b53b10ea61 Fix Train LoRA crash when training_dtype is "none" with bfloat16 LoRA weights (#13145)
7d5534d8e5 feat(assets): register output files as assets after prompt execution (#12812)
5ebb0c2e0b FP8 bwd training (#13121)
Compare 10 commits »
wangbo synced commits to refs/pull/13048/merge at wangbo/ComfyUI from mirror 2026-03-26 11:34:35 +08:00
9a431ba973 Merge fbc3b0fed3f873cfb813499e63548db61377765f into 3b15651bc6
3b15651bc6 bump manager version to 4.1 (#13156)
a55835f10c fix(api-nodes): made Reve node price badges more precise (#13154)
b53b10ea61 Fix Train LoRA crash when training_dtype is "none" with bfloat16 LoRA weights (#13145)
7d5534d8e5 feat(assets): register output files as assets after prompt execution (#12812)
Compare 9 commits »
wangbo synced commits to refs/pull/12413/merge at wangbo/ComfyUI from mirror 2026-03-26 11:34:35 +08:00
ca253024d2 Merge 97ba2a5155 into a55835f10c
a55835f10c fix(api-nodes): made Reve node price badges more precise (#13154)
b53b10ea61 Fix Train LoRA crash when training_dtype is "none" with bfloat16 LoRA weights (#13145)
7d5534d8e5 feat(assets): register output files as assets after prompt execution (#12812)
5ebb0c2e0b FP8 bwd training (#13121)
Compare 5 commits »