Enderfga
fba7170ed0
[feat] Thread optional r_timestep through WanTransformer3DModel.forward
...
The forward signature now accepts r_timestep as an explicit kwarg (not
silently swallowed by **kwargs), gets flattened in parallel with timestep
when the input is 2-D (Wan 2.2 ti2v case), and is forwarded to the
condition_embedder.
The constructor now propagates the four arch_config flags
(r_embedder/r_embedder_fusion/r_embedder_gate_value/r_embedder_deltatime_type)
into the WanTimeTextImageEmbedding so a YAML can opt into AnyFlow's
dual-timestep conditioning through training.dit_config overrides.
Defaults preserve byte-identity with the legacy single-timestep path —
when r_timestep is omitted the delta_embedder branch is skipped entirely
(test_embedder_enabled_without_r_timestep_is_bit_identical_to_legacy).
2026-07-11 18:36:14 -07:00
Enderfga
0e042101f5
[feat] Extend WanTimeTextImageEmbedding with optional r_timestep
...
When r_embedder=True, allocates a delta_embedder (deep-copied from
time_embedder for matching initialization, per the AnyFlow reference's
setup_flowmap_model() pattern) and registers a non-persistent gate buffer.
Forward now accepts an optional r_timestep:
- additive (default): temb_t + g * delta_emb
- gated: (1 - g) * temb_t + g * delta_emb
Both branches are gated by r_embedder=True AND r_timestep is not None,
so existing call sites that don't pass r_timestep stay byte-identical
(verified by test_embedder_enabled_without_r_timestep_is_bit_identical_to_legacy).
The gate is a non-persistent buffer — it's a hyperparameter, not a
learned weight, and stays out of state_dict so checkpoints remain
portable across gate values.
deltatime_type controls whether the delta_embedder consumes r directly
(default, matching AnyFlow's deltatime_type='r') or (t - r).
2026-07-11 18:36:14 -07:00
Enderfga
836332786b
[feat] Add r_embedder config flags to Wan (defaults preserve bit-identity)
...
Adds four fields to WanVideoArchConfig for AnyFlow dual-timestep conditioning:
- r_embedder: bool — enables the delta_embedder allocation
- r_embedder_fusion: 'additive' (default) or 'gated'
- r_embedder_gate_value: float — gate g in gated fusion
- r_embedder_deltatime_type: 'r' (default) or 't-r'
Also adds two regex entries to param_names_mapping so HF AnyFlow checkpoints
load delta_embedder weights into the FastVideo-internal mlp.fc_in/fc_out layout.
Both regex are no-ops on plain Wan checkpoints.
The defaults preserve bit-identity with every existing Wan-based method
(DMD2, Self-Forcing, KD, DFSFT) — no delta_embedder is allocated and no
extra computation runs on the embedder forward path.
2026-07-11 18:36:14 -07:00
Raghav K
055586703d
[perf]: register a real backward for FA2 default + masked/varlen custom ops (training-under-compile) ( #1388 )
2026-07-12 00:45:27 +00:00
Mac Lee
5d89f86675
[ci] Stop forcing FA4 in model-load lanes ( #1561 )
2026-07-11 14:10:36 -07:00
Satyam Srivastava
19a51a1fe6
[ci] Trigger performance benchmarks for performance code changes ( #1583 )
2026-07-10 20:21:33 -07:00
William Lin
d3232cea5a
[ci]: gate the full-suite trigger on pre-commit and docs build ( #1572 )
2026-07-11 02:56:49 +00:00
Raghav K and SolitaryThinker
0c90c8c24d
[bugfix] nvfp4: cast fp32 inputs to bf16 instead of asserting ( #1488 )
...
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com >
2026-07-10 21:39:17 +00:00
Mingjia Huo and Claude Fable 5
4c08ffce49
[feat] World model training using third person games ( #1443 )
...
Co-authored-by: Claude Fable 5 <noreply@anthropic.com >
2026-07-10 05:09:07 +00:00
Atharv Ramesh
af4a77553c
[ci]: add SSIM reference bootstrap flow ( #1522 ) ( #1547 )
2026-07-10 01:49:38 +00:00
alexzms and SolitaryThinker
c096fda1eb
[docs] Add LTX-2.3 distilled inference run configs (t2v/i2v × 5+2/8+3 × resolutions) ( #1568 )
...
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com >
2026-07-09 18:53:12 +00:00
William Lin
8f47e85be0
[bugfix]: retry remote image downloads in load_image ( #1570 )
2026-07-09 07:33:24 -07:00
William Lin
90d3bd19eb
[infra] Deliver per-job Buildkite env to Modal CI at runtime, not as image layers ( #1569 )
2026-07-09 06:33:37 -07:00
Mac Lee and SolitaryThinker
afb4f7d3c5
[ci]: emit v2 performance result schema ( #1551 )
...
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com >
2026-07-09 06:11:15 +00:00
02e1143f22
[feat] Add Kandinsky-5 T2V/I2V pipeline support ( #1471 )
...
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com >
Co-authored-by: leffff <levnovitskiy@gmail.com >
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com >
2026-07-07 14:43:30 -07:00
Kared and SolitaryThinker
e2f4d1a7b5
[feat]: add SwanLab tracker ( #1461 )
...
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com >
2026-07-07 08:35:38 +00:00
Kaiqin Kong and SolitaryThinker
f037351146
[feat] Add Clean-history Teacher Forcing and Causal Consistency Distillation ( #1505 )
...
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com >
2026-07-07 08:13:11 +00:00
Mac Lee and SolitaryThinker
1ee11e08dc
[ci]: add performance fingerprint cohorts ( #1546 )
...
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com >
2026-07-07 07:24:00 +00:00
595f0ea60e
[feat] Add DreamX-World 5B Cam and AR pipelines ( #1538 )
...
Co-authored-by: Suckl <Suckl@users.noreply.github.com >
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com >
2026-07-07 06:16:18 +00:00
William Lin
d921832cd2
[misc]: reserve CPU/memory for timing-sensitive Modal CI lanes ( #1566 )
2026-07-06 22:13:53 -07:00
William Lin
629697629a
[bugfix]: free CUDA memory between train-framework model tests (LongCat OOM on L40S) ( #1565 )
2026-07-06 21:27:30 -07:00
William Lin
a25313beec
[ci]: wire fastvideo/tests/ops/ into the unit-test lane ( #1559 )
2026-07-06 12:07:26 -07:00
William Lin
dbde64385b
[bugfix]: bump FA4 pin to the CuTe DSL 4.6 compatible rev ( #1564 )
2026-07-06 12:06:59 -07:00
William Lin
9d909f5f04
[test]: remove dead and duplicate tests (-489 lines) ( #1556 )
2026-07-05 15:53:40 -07:00
William Lin
76b0550c15
[ci]: run pre-commit on fork PRs without manual approval ( #1555 )
2026-07-05 14:18:16 -07:00
William Lin
384c1e9493
[misc]: update reseed-performance-baseline skill for the hf_store move ( #1545 follow-up) ( #1553 )
2026-07-05 14:16:55 -07:00
William Lin
b1dbcc93f6
[misc]: reformat fastvideo/performance to the repo yapf config ( #1554 )
2026-07-05 14:16:20 -07:00
William Lin
b93833772e
[ci]: guard against test directories no CI lane collects ( #1552 )
2026-07-05 14:07:38 -07:00
Mac Lee
30b523edd6
[ci] Normalize performance stage component metrics ( #1475 ) ( #1550 )
2026-07-05 14:05:26 -07:00
Mac Lee and SolitaryThinker
6aab7f3832
[ci] cover Hunyuan 1.5 chat-list text preprocessing ( #1518 )
...
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com >
2026-07-05 12:05:25 -07:00
Mac Lee
9cd53fe5f8
[ci] Add metric-specific performance thresholds ( #1545 )
2026-07-05 12:04:33 -07:00
Mac Lee
6a32cf3a5e
[ci]: expose LoRA extraction slash command ( #1542 )
2026-07-05 06:45:31 -07:00
William Lin
98be9b3da2
[bugfix]: address the three remaining #1447 review findings ( #1549 )
2026-07-05 06:43:27 -07:00
Mac Lee
c53e85b767
[ci] Add v2 performance benchmark config identity fields ( #1544 )
2026-07-05 06:18:15 -07:00
William Lin
40a8bd2d3b
[bugfix]: skip ThunderKittens kernels on aarch64 and document the kernel build matrix ( #1548 )
2026-07-05 06:13:11 -07:00
Mac Lee and SolitaryThinker
31aa115611
[bugfix]: preserve FSDP hooks for RMSNorm qk norms ( #1513 )
...
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com >
2026-07-05 04:15:49 +00:00
zainnh and SolitaryThinker
98ac10a528
[infra] Auto-rebuild CUDA images when docker/Dockerfile changes on main ( #1526 )
...
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com >
2026-07-04 16:25:46 -07:00
William Lin
a5a6d171e5
[attn] Make FA4 explicit opt-in via FASTVIDEO_FA4 and delete the runtime fallback machinery ( #1540 )
2026-07-04 14:51:00 -07:00
William Lin
51ed1ea423
[build] Bump fastvideo-kernel pin to 0.3.2 ( #1541 )
2026-07-03 14:48:15 -07:00
Mac Lee
00ec3e7388
[bugfix]: compute VSA topk from padded blocks ( #1517 )
2026-07-03 14:16:51 -07:00
William Lin
0d626ef2d1
[kernel] Bump fastvideo-kernel pin to 0.3.1 and version the FA4 tile_mn port as 0.3.2 ( #1539 )
2026-07-03 14:11:08 -07:00
William Lin
b36d0ef085
[infra] Add DGX Spark and multi-architecture CUDA support ( #1447 )
2026-07-03 13:41:43 -07:00
William Lin
10c8c5df4d
[misc] reorg: relocate ui/ and performance_dashboard/ under apps/ ( #1537 )
2026-07-02 20:29:17 -07:00
William Lin
31e26abec4
[chore] release fastvideo-kernel 0.3.1 ( #1520 )
2026-06-30 12:25:27 -07:00
sumyyyyy and SolitaryThinker
9e83ba630c
[kernel] Extract VSA utility functions into fastvideo_kernel ( #1408 )
...
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com >
2026-06-30 12:19:25 -07:00
William Lin
fc02a9ce8e
[ci] aarch64 kernel wheel: build for Blackwell (sm_100a/sm_120a), not Hopper ( #1516 )
2026-06-29 16:10:42 -07:00
William Lin
3d36160fc4
[bugfix] -fsigned-char so ThunderKittens compiles on aarch64 (Grace Hopper) ( #1515 )
2026-06-29 14:05:38 -07:00
William Lin
a11ec43de2
[ci] build + publish aarch64 (Grace Hopper) kernel wheels alongside x86_64 ( #1514 )
2026-06-29 12:28:34 -07:00
William Lin
2656d6530c
[bugfix] compile FP4 (attn_qat_infer) kernels for sm_120a only via per-arch split ( #1508 )
2026-06-29 11:07:27 -07:00
Kevin Lin
e3f54e7169
[bugfix] Fix causal attention mask for Blackwell FP4 MMA column layout ( #1506 )
2026-06-28 20:36:40 -07:00
Enderfga