nodes/hao-ai-lab-FastVideo
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
1,204 Commits 344 Branches 8 Tags
866a3cdb91a85aef8561847365fc8564add9e4c8
Commit Graph
835 Commits
Author SHA1 Message Date
Kyle Hu ce970eb715 [perf]: training-free token reduction for MiniMax H3 (#2004) 2026-11-08 04:06:38 +00:00
Arunkumar EliandAryan Kumar edd517e3fa [bugfix] Make torch.compile reduce-overhead work in the shared denoising loop (Wan T2V) (#2047) 
Co-authored-by: Aryan Kumar <156166681+aryan5v@users.noreply.github.com>
 2026-11-07 23:55:36 +00:00
Mingyang ZhongandClaude Opus 5.5 a01c50a806 [bugfix] End partial causal Wan mask blocks at the real token boundary (#1976) 
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
 2026-11-06 02:38:27 +00:00
Raghav KClaude Opus 5.5Aryan Kumar
 		0b3d2db9f3 [bugfix] Stop the preprocessing dataset from keeping every decoded video in memory (#2069) 
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Aryan Kumar <156166681+aryan5v@users.noreply.github.com>
 2026-11-06 02:31:54 +00:00
Keith 58940992ae [bugfix] Streaming: answer malformed client frames with invalid_message instead of dropping the session (#2073) 2026-11-05 18:14:53 +00:00
Raghav KandClaude Opus 5.5 f8864246db [bugfix] Map merged-LoRA export keys back to HF names when LoRA was enabled after loading (#2070) 
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
 2026-11-05 15:40:06 +00:00
Raghav KandClaude Opus 5.5 47010cf91c [bugfix] Fall back to PyAV when torchvision.io.read_video is unavailable in preprocessing (#2068) 
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
 2026-11-05 13:26:45 +00:00
Minshen ZhangandClaude Opus 5.5 764f50c711 [feat] OpenAI video API: accept data: URLs for video and audio references (#2030) 
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
 2026-11-05 04:55:07 +00:00
FV SuperAgentandAryan Kumar d866c2d507 [bugfix] Gate the FA4 NVFP4 path on the installed kernel, not the import (#2067) 
Co-authored-by: Aryan Kumar <156166681+aryan5v@users.noreply.github.com>
 2026-11-04 19:50:30 -08:00
Aryan KumarandAryan Kumar c845a8dee2 [feat] Per-request community LoRA in the OpenAI video API (#2026) 
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com>
 2026-11-04 19:44:47 -08:00
tianwei-moandClaude Opus 5.5 36dc971dc6 [feat] MiniMax-H3 Ref2VA: continue an earlier video from its last frames (#2001) 
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
 2026-11-05 01:24:11 +00:00
tianwei-moandClaude Opus 5.5 229c43c8d3 [feat] MiniMax-H3 Ref2VA SFT: train each clip at its own canvas and frame count (#2003) 
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
 2026-11-05 01:12:34 +00:00
Haoyu Cheng 9e1915ceb4 [bugfix] Honor output_video_name when selecting output filenames (#2044) 2026-11-04 23:25:14 +00:00
Kostiantyn Dementiev 40651713d5 [bugfix] Count the wait for a concurrency slot in bench latency (#1960) 2026-11-04 23:14:33 +00:00
li-lizhearyan5vAryan Kumar
 		702e62f2b2 [bugfix] SP warmup: use the rank's own device instead of hardcoded CUDA (#1988) 
Signed-off-by: li-lizhe <147392333@qq.com>
Co-authored-by: aryan5v <aryan5v@users.noreply.github.com>
Co-authored-by: Aryan Kumar <156166681+aryan5v@users.noreply.github.com>
 2026-11-04 20:28:49 +00:00
Aryan KumarandAryan Kumar 4b4df1a94e [perf] Fused Q/K preconditioning + NVFP4 quantize for FA4 NVFP4 attention (bit-identical) (#2052) 
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com>
 2026-11-03 01:07:11 -08:00
Aryan KumarandAryan Kumar 7066408fb3 [feat] Pika H3 Flex NVFP4 QAD training: calibration, QAD, FP4 attention and export (#2051) 
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com>
 2026-11-03 01:04:26 -08:00
Aryan KumarandAryan Kumar fcec554456 [feat]: shared FastH3 NVFP4 FP4-attention QAD recipe (OmniRef + V2 T2AV) (#1989) 
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com>
 2026-11-03 00:54:28 -08:00
Aryan KumarandAryan Kumar 3fe9c0c70d [bugfix]: get the full CI suite green on main (pytest 9 flag teardown, Qwen golden env) (#2059) 
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com>
 2026-11-02 23:32:40 -08:00
William Lin de873eae2d [bugfix] Pass disable_autocast through to FastVideoArgs (#2056) 2026-11-02 13:42:08 -08:00
Aryan KumarAryan KumarAryan
 		4f80c5a3da [feat] Open-H3-Regenerate-2K: regenerate FastH3 QAD, Trim and V2 drafts at 2K (text to 2K 3.1-3.4x faster) (#2037) 
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com>
Co-authored-by: Aryan <aryan@aryan5v.dev>
 2026-11-02 03:44:28 -08:00
Aryan KumarandAryan Kumar b0047cfedc [feat] Open-H3-Regenerate-2K: NVFP4 H3 video decoder (decode 23 s to 5 s at 2K) (#2036) 
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com>
 2026-11-01 22:51:41 -08:00
Aryan KumarandAryan Kumar ac81198d33 [new-model] Open-H3-Regenerate-2K: 768P H3 draft to 2560x1440, served on the OpenAI video routes (#2035) 
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com>
 2026-11-01 22:45:09 -08:00
Aryan KumarandAryan Kumar cb02062930 [new-model] Add Pika H3 Flex inference support (#2038) 
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com>
 2026-11-01 22:33:10 -08:00
Aryan KumarandAryan Kumar d954fad9ec [perf] MiniMax H3: faster FP4 attention padding, uint8 VAE pixel path and cuDNN FP4 GEMMs (#2046) 
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com>
 2026-11-01 22:23:10 -08:00
Aryan KumarandAryan Kumar 166b9b8a07 [perf] MiniMax H3: fuse VAE decoder QK RMSNorm and partial RoPE (#2041) 
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com>
 2026-11-01 22:15:35 -08:00
Aryan KumarandAryan Kumar 7e1ed5e28e [perf] MiniMax H3: decode eight VAE spatial tiles per call by default (#2040) 
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com>
 2026-11-01 22:06:23 -08:00
Aryan KumarandAryan Kumar cb6a9a13e1 [bugfix] FA4 JIT cache vs tuning env, MXFP8 import and scale dtype, H3 VSA gate double quantization (#2049) 
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com>
 2026-11-01 19:11:25 -08:00
Aryan KumarandAryan Kumar 0d1244a869 [perf] MiniMax H3: opt-in faster kernel tier for the DiT fused ops (#2042) 
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com>
 2026-11-01 15:22:09 -08:00
Aryan KumarandAryan Kumar 18b2a35106 [bugfix] FA4: stop forwarding max_seqlen to the varlen kernel (#2043) 
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com>
 2026-11-01 15:21:38 -08:00
Aryan Kumar 7203b4976e [bugfix]: compile only the LoRA delta, not the wrapped base layer 
BaseLayerWithLoRA.forward was decorated with torch.compile, so wrapping a linear also compiled its base layer.
Compiling a packed NVFP4/FP8 linear on its own changes its numerics, so a wrapped layer at strength 0 could differ
from the unwrapped layer. forward now adds no compiled region: the base layer runs exactly as it would unwrapped
(eagerly, or traced with a caller's compiled block), and only the adapter delta is compiled.
 2026-10-28 14:49:15 -07:00
Aryan Kumar 6bb444eef1 [bugfix]: run the base layer while a LoRA wrapper has no adapter 2026-10-28 14:49:15 -07:00
Aryan Kumar 6743f4b4b7 Revert "[new-model] Add PAN2 text-to-video and image-to-video" (#2025) 2026-10-27 18:03:59 -07:00
zoechen281WeiersAryan Kumararyan5v
 		465654d900 [bugfix] Reject missing CFG negative conditioning during input validation (#2011) 
Co-authored-by: Weiers <16632241+zoechen1212@user.noreply.gitee.com>
Co-authored-by: Aryan Kumar <156166681+aryan5v@users.noreply.github.com>
Co-authored-by: aryan5v <aryan5v@users.noreply.github.com>
 2026-10-27 08:45:21 +00:00
Jack LiangClaude Opus 5.5Aryan Kumar
 		69b3829293 [bugfix] Normalize local and HF-cache model paths before registry lookup (#1997) 
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Aryan Kumar <156166681+aryan5v@users.noreply.github.com>
 2026-10-27 07:47:20 +00:00
William Linandaryan5v a0e6cf1f50 [new-model] Add PAN2 text-to-video and image-to-video (#1961) 
Co-authored-by: aryan5v <email.aryan1005@gmail.com>
 2026-10-25 20:20:00 -07:00
Loay RashidandAryan Kumar 0d05b09c57 [new-model] Add initial Qwen-Image-2.1 inference (#1918) 
Co-authored-by: Aryan Kumar <156166681+aryan5v@users.noreply.github.com>
 2026-10-25 17:28:49 +00:00
posiposi443-opsandAryan Kumar 8db2d71ef2 [perf] Cache MLX H3 text refinement and speed up attention (#2002) 
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com>
 2026-10-23 12:40:47 -07:00
Aryan Kumar 9bbdd18548 [ci]: Fix env-policy and schema-parity contract tests 
The unit lane fails on main in two contract tests. All three failures are
bookkeeping that drifted from the code, not runtime bugs:

- `test_env_policy.py::test_env_doc_table_matches_registry`: the generated
  table in docs/contributing/env_vars.md is stale for
  FASTVIDEO_MLX_QUANT_GEMM. Regenerated with the documented command
  (`python fastvideo/tests/contract/test_env_policy.py`).
- `test_env_policy.py::test_env_access_follows_policy`: two new violations.
  - fastvideo/tests/mlx/test_mlx_affine_dq_gemm.py wrote
    FASTVIDEO_MLX_QUANT_GEMM through monkeypatch.setenv; now uses the
    registered field's override() through the env_overrides fixture.
  - fastvideo/tests/train/models/test_load_kandinsky5.py read
    FASTVIDEO_ATTENTION_BACKEND with `not in os.environ`; now uses
    envs.FASTVIDEO_ATTENTION_BACKEND.is_set() and override(). Its entry in
    KNOWN_VIOLATIONS is deleted, as the shrinking-list rule requires.
- `test_schema_parity_inventory.py::..._extension_fields_are_classified`:
  MiniMaxH3PipelineConfig declares video_scheduler_shift and
  audio_scheduler_shift (training-time noise clocks rejected by
  check_pipeline_config at inference); both are added to
  pipeline_config_extensions.internal_only.

Test plan (CPU, no GPU required):

  python -m pytest fastvideo/tests/contract/test_env_policy.py \
      fastvideo/tests/api/test_schema_parity_inventory.py \
      fastvideo/tests/mlx/test_mlx_affine_dq_gemm.py -q
  # 45 passed

  python -m pytest fastvideo/tests/contract/ fastvideo/tests/api/ -q -m "not gpu"
  # 501 passed
 2026-10-23 11:08:37 -07:00
Aryan KumarandAryan Kumar 7c2b447264 [ci] Run single-GPU CI lanes on a DGX Spark runner; weekly full suite on main (#2007) 
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com>
 2026-10-22 13:44:55 -07:00
William Lin e136c2b7f8 [infra]: fix the DGX Spark (GB10) Docker image: importable flashinfer, sm_121a kernels, native-matching output (#1986) 2026-10-22 11:59:31 -07:00
Aryan KumarandAryan Kumar fa2d1324c6 [feat]: FastH3 Omni (Ref2VA) on Apple Silicon MLX (#2000) 
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com>
 2026-10-21 12:01:54 -07:00
William Lin cdda1b7535 [feat] Add H3 PDD/DMPDD training and packed Ref2VA preprocessing (#1984) 2026-10-20 11:15:57 -07:00
Aryan KumarandAryan Kumar 089a5a55a0 [feat] Cookbook serving, hardware catalog, and landing page (#1999) 
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com>
 2026-10-20 09:10:29 -07:00
Aryan KumarAryan KumarClaude Opus 5.5
 		376e445f0e [feat]: FastH3 OmniRef inference on consumer devices (RTX 5090 / PRO 6000 / DGX Spark / 4090) (#1980) 
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
 2026-10-19 14:57:36 -07:00
Aryan KumarAryan KumarSolitaryThinker
 		2099dd5533 [feat]: Community MiniMax-H3 LoRA loading with quantized side branch (#1973) 
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com>
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com>
 2026-10-19 14:29:23 -07:00
Aryan KumarAryan Kumargreptile-apps[bot]
 		0033a9b11d [bugfix] Fix FastH3 regional compile recompiling for every prompt length (#1942) 
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com>
Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>
 2026-10-18 18:22:06 -07:00
Aryan Kumargemini-code-assist[bot]SolitaryThinker
 		91a1c50909 [fix] Fix latent handoff and FastWan-QAD example path (#1557) 
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com>
 2026-10-18 17:36:40 -07:00
lpc0220rootClaude Opus 5.5root
 		3658b120da [bugfix] Fix tests that fail on main: FastH3 Trim Spark audio, test isolation (#1957) 
Co-authored-by: root <root@ptyche0124.ptyche.clusters.nvidia.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: root <root@ptyche0040.ptyche.clusters.nvidia.com>
 2026-10-18 16:07:59 -07:00
lpc0220androot d4ed61cc7c [refactor] VSA: rename the sm100a block-sparse kernel to ptx_blackwell (#1956) 
Co-authored-by: root <root@ptyche0124.ptyche.clusters.nvidia.com>
 2026-10-17 21:10:19 -07:00