alexzms
5671ebc718
[perf] LTX-2.3 NVFP4 bench: VAE-tiling / latent-output knobs + long-token docs
...
Add LTX23_VAE_TILING, LTX23_DECODE (output_type=latent to skip VAE decode),
and LTX23_WARMUP/LTX23_MEASURED env knobs to the t2v bench harness so long /
high-res runs (241/481 frames) can be measured. output_type=latent lets us
time the DiT (denoise/refine) cleanly without the slow CPU-bound tiled VAE
decode dominating wall-time.
Update INSTRUCTION_ltx23_nvfp4.md: document the new knobs, add a long-token
scaling section (121/241/481-frame nvfp4-vs-bf16 numbers; refine attention N²
crossover), and correct the tiling/decode notes.
2026-06-15 16:55:13 -07:00
alexzms
57a2fa9ec1
[perf] LTX-2.3 distilled NVFP4 + FA4-FP4 + CUDA-graphs benchmark on GB300
...
Make compile + FA4-FP4 attention work and add a parametrized speed harness.
- flash_attn.py: wrap `_nvfp4_quantize_for_fa4` as a torch.library.custom_op
+ register_fake so torch.compile/dynamo treats the flashinfer FP4 Q/K quant
as an opaque leaf (can't trace its functools.cache lock / JIT subprocess);
pre-build the FP4 quant module in FlashAttentionImpl.__init__ to warm the
cache before compile. Required for compile + nvfp4_fa4 to run at all.
- bench_ltx2_3_distilled_t2v.py: parametrized t2v harness (compile / nvfp4 /
fa4 / cudagraph mode / resolution / frames / steps via env) with per-stage
timing.
- basic_ltx2_3_distilled_i2v_{uncompiled,compiled}{,_nvfp4}.py: i2v variants.
- INSTRUCTION_ltx23_nvfp4.md: full environment + command-line reproduction.
Result (t2v 832x1280, 8+3, GB300): nvfp4+FA4+cudagraphs = 3.00s e2e, the
fastest config (vs bf16+FA4+cudagraphs 3.81s, nvfp4+FA4 default 6.35s). The
NVFP4 denoise penalty is per-step launch-bound; CUDA graphs removes it.
2026-06-14 11:57:10 -07:00
Junda Su and Peiyuan Zhang
5854aec2ce
[feat] Add Wan RL DiffusionNFT training ( #1450 )
...
Co-authored-by: Peiyuan Zhang <a1286225768@gmail.com >
2026-06-11 21:18:59 -07:00
alexzms
2a4fe697a6
[docs] LTX-2.3 distilled i2v: typed-API example (from_config + generate) ( #1448 )
2026-06-10 10:58:14 -07:00
alexzms
921db7479d
[perf] LTX-2.3 distilled i2v: drop max-autotune from compile kwargs ( #1445 )
2026-06-10 10:06:41 -07:00
Aryan Kumar and Aryan Kumar
7f539424cb
[feat]: add Lucy Edit inference scaffold ( #1363 )
...
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com >
2026-06-09 15:21:43 -07:00
d922ab2cbc
[model] Flux2 Klein Port ( #1349 )
...
Co-authored-by: Gnav3852 <63612880+Gnav3852@users.noreply.github.com >
Co-authored-by: Mac Lee <macthecadillac@gmail.com >
2026-06-09 14:55:55 -07:00
Kaiqin Kong and mergify[bot]
efcc245c2e
[feat] VLM as judge for WM ( #1429 )
...
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2026-06-05 04:25:54 +00:00
alexzms and mergify[bot]
922e7e0813
[docs] LTX-2.3 distilled i2v example with compile + timing breakdown ( #1430 )
...
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2026-06-05 01:26:50 +00:00
Kaiqin Kong and SolitaryThinker
afdb6fbfa5
[feat] Add MatrixGame3.0 ( #1201 )
...
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com >
2026-05-27 21:07:01 +00:00
Shao Duan
36682797a0
[feat] eval: input ergonomics + Evaluator features + bug fixes ( #1392 )
2026-05-26 13:59:45 -07:00
Shao Duan and abaghyangor
321d5112b4
[refactor] eval: consolidate FVD into common.fvd, remove benchmarks/fvd ( #1380 )
...
Co-authored-by: abaghyangor <abaghyangor@gmail.com >
2026-05-24 12:09:47 -07:00
Raghav K
68179cd752
[docs] Document enable_torch_compile (+ A/B example) ( #1366 )
2026-05-22 00:25:42 -07:00
e8597b7448
[Bugfix] FP4 FA4 installation fix ( #1367 )
...
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com >
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com >
2026-05-19 22:45:24 -07:00
Junda Su
72cb427cd9
[feat]: add FastLTX-2.3 Gradio demo package (draft) ( #1247 )
2026-05-17 23:03:16 -07:00
Mingjia Huo
63030cf6ec
[fix] Fix causal self-forcing attention settings ( #1355 )
2026-05-17 22:59:47 -07:00
Kaiqin Kong and mergify[bot]
773d44b875
[misc] Rename MatrixGame to MatrixGame2 ( #1357 )
...
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2026-05-17 22:57:03 -07:00
1df513922f
[feat] Add minimal LoRA finetuning support to the YAML training stack ( #1242 )
...
Co-authored-by: alexzms <3036648523@qq.com >
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2026-05-17 14:49:13 -07:00
Shao Duan and klhhhhh
6b2c731596
[feat] eval: add audio metrics ( #1352 )
...
Co-authored-by: klhhhhh <1412841649@qq.com >
2026-05-16 14:47:37 -07:00
d2ffec5cce
[perf] Dreamverse 14/14: Add LTX2 profile speedups ( #1337 )
...
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com >
Co-authored-by: Matthew Noto <99706358+RandNMR73@users.noreply.github.com >
Co-authored-by: XOR-op <17672363+XOR-op@users.noreply.github.com >
Co-authored-by: Zhang Peiyuan <42993249+jzhang38@users.noreply.github.com >
2026-05-15 18:10:47 -07:00
Shao Duan
17f07bc313
[feat] eval: async VideoPool + metric streamlines ( #1320 )
2026-05-13 15:57:37 -07:00
William Lin
325861fb99
[misc]: PR-1225 sync — housekeeping (1/12) ( #1347 )
2026-05-13 15:21:52 -07:00
0403c6f47e
[infra] Dreamverse 09/14: Add Docker and launch scripts ( #1332 )
...
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com >
Co-authored-by: Matthew Noto <99706358+RandNMR73@users.noreply.github.com >
Co-authored-by: XOR-op <17672363+XOR-op@users.noreply.github.com >
Co-authored-by: Zhang Peiyuan <42993249+jzhang38@users.noreply.github.com >
2026-05-12 18:01:11 -07:00
b4458e5bae
feat: FP4 Flash Attention 4 for Blackwell GPUs ( #1221 )
...
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com >
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2026-05-12 17:04:39 -07:00
Kaiqin Kong and mergify[bot]
a790153705
[bugfix] MatrixGame2 SF distillation under gradient checkpointing ( #1340 )
...
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2026-05-12 17:01:04 -07:00
Raghav K
9d0af28307
[feat] Add Cosmos 2.5 T2W training pipeline (LoRA + full fine-tune) ( #1227 )
2026-05-11 19:15:00 -07:00
d6dfe95466
[feat] FastVideo World Model Training ( #1179 )
...
Co-authored-by: mignonjia <mhuo@ucsd.edu >
Co-authored-by: alexzms <3036648523@qq.com >
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2026-05-11 05:13:12 +00:00
f633e30ebb
[feat]: add LongCat bidirectional finetuning support ( #1244 )
...
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com >
Co-authored-by: alexzms <3036648523@qq.com >
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2026-05-09 12:20:43 -07:00
Shao Duan
f6396fb8c6
[feat] Add fastvideo.eval video evaluation suite ( #1305 )
2026-05-07 20:31:15 -07:00
William Lin
140bd1a6cf
[misc]: standardize install instructions on uv pip install ( #1279 )
2026-05-02 12:45:50 -07:00
William Lin
c77a76c6af
[feat] Stable Audio Open 1.0: T2A + A2A + RePaint inpainting (native) ( #1260 )
2026-05-01 00:07:11 -07:00
alexzms
d14d5aadea
[feat] Cosmos 2.5 training support in fastvideo.train ( #1224 )
2026-05-01 01:15:02 +00:00
William Lin
145a3f166b
[feat] [4/n] Improve API: refactor sampling param and merge with presets ( #1234 )
2026-04-16 14:10:02 -07:00
William Lin
c591d6d2a6
[feat] [2/n] Improve API: add initial support in video_generator ( #1220 )
2026-04-06 10:33:54 -07:00
vishruthb
5789955bbe
[feat] add gen3c (cosmos-7b) model and pipeline support ( #1059 )
2026-04-01 11:42:02 +00:00
alexzms
2085a4fc4a
[bugfix]: fix VAE temporal tiling blend corruption in tiled_encode ( #1181 )
2026-03-29 23:12:42 +00:00
alexzms
30129a3350
[misc]: reorganize training configs and add documentation ( #1177 )
2026-03-26 16:38:31 -07:00
alexzms
71bfc13d75
[feat]: add HunyuanVideo model plugin for fastvideo/train framework ( #1175 )
2026-03-24 16:49:51 -07:00
alexzms
14261bc98c
[feat] pre-commit support 120 col num ( #1167 )
2026-03-11 20:19:30 -07:00
alexzms
d92858659d
[feat] Self-Forcing methods in refactored training infra ( #1164 )
2026-03-09 20:20:59 -07:00
1a383f3f66
[refactor] train v1 clean up
...
Co-authored-by: alexzms <3036648523@qq.com >
Co-authored-by: Peiyuan Zhang <a1286225768@slurm-h200-204-215.slurm-compute.tenant-slurm.svc.cluster.local >
2026-03-09 18:58:56 -07:00
alexzms and Peiyuan Zhang
bc27a032c5
[feat] Refactor training framework into fastvideo/train ( #1159 )
...
Co-authored-by: Peiyuan Zhang <a1286225768@gmail.com >
2026-03-09 15:16:42 -07:00
Zhang Peiyuan
cd1b7cf139
[Refactor] SP Mask --> original seq len; HunyuanVideo 1.5 does not need mask ( #1142 )
2026-03-04 11:44:47 +08:00
Zhang Peiyuan
fea556269b
[Misc] Fix memory leakage in VideoGenerator ( #1132 )
2026-02-26 19:51:32 -08:00
Jinzhe Pan
5433f6e80b
[fix] preprocessing issue ( #1134 )
2026-02-25 21:49:52 -08:00
Zhang Peiyuan and Will Lin
fc80d95d7e
[Misc] Remove STA ( #1124 )
...
Co-authored-by: Will Lin <wlsaidhi@gmail.com >
2026-02-23 15:14:42 -08:00
Zhang Peiyuan
128d0b7fc5
[Misc] Remove Teacache ( #1121 )
2026-02-22 16:53:07 -08:00
10fc92dba5
Upstream LTX2 Training ( #1116 )
...
Co-authored-by: RandNMR73 <notomatthew31@gmail.com >
Co-authored-by: JerryZhou54 <zhouw.jerry2017@outlook.com >
Co-authored-by: Davids048 <jundasu@ucsd.edu >
Co-authored-by: Peiyuan Zhang <a1286225768@slurm-h200-204-239.slurm-compute.tenant-slurm.svc.cluster.local >
2026-02-21 16:06:55 -08:00
Zhang Peiyuan and gemini-code-assist[bot]
76d3ba69e0
[Misc] clean up VSA finetuning examples. ( #1111 )
...
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
2026-02-18 11:37:20 -08:00
8e39ce38c9
[Feat] Native dit implementation for SD3.5 ( #1093 )
...
Co-authored-by: Ishan Vaish <vaish.ishan@gmail.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-02-18 10:19:40 +08:00
alexzms