d8ca702dfd
[bugfix] registry: register magi_human presets ( #1751 )
...
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com >
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com >
2026-10-07 01:28:53 -07:00
William Lin
33d81730e9
[feat]: Kandinsky 6 Lite: register the 3.2B checkpoints and add docs + cookbook recipes ( #1927 )
2026-10-06 17:09:53 -07:00
02a027c49f
[feat]: FastH3 on DGX Spark and Apple Silicon ( #1920 )
...
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com >
Co-authored-by: Aryan Kumar <aryank@Aryans-Mac-Studio.local >
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com >
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com >
2026-10-06 12:17:11 -07:00
Aryan Kumar
e235b6a332
[docs] Announce FastH3 on consumer hardware and FastH3 Trim ( #1923 )
2026-10-06 11:39:24 -07:00
8b51380466
[feat]: FastH3 on single RTX GPUs ( #1919 )
...
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com >
Co-authored-by: Aryan Kumar <aryank@Aryans-Mac-Studio.local >
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com >
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com >
2026-10-06 11:37:01 -07:00
William Lin
527a3165b9
[docs]: add Kandinsky 6 to the README news ( #1925 )
2026-10-06 11:05:40 -07:00
f06d824054
[feat]: add Kandinsky-6.0 TI2VA and VSR pipelines ( #1924 )
...
Co-authored-by: leffff <levnovitskiy@gmail.com >
Co-authored-by: Kirill <kykozlov@gmail.com >
Co-authored-by: shaoxiongduan <shaoxiongduan@gmail.com >
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com >
2026-10-06 10:50:17 -07:00
Kyle Hu
6ded84ee0f
[ci]: improve CI efficiency by reordering lanes ( #1914 )
2026-10-04 12:35:12 -07:00
a4d6416c52
[docs] Add OpenAI serving recipes for Wan cookbook pages ( #1906 )
...
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com >
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com >
2026-10-04 12:20:06 -07:00
lpc0220 and Claude Opus 5.5
0cc41a22dc
[bugfix] sm_100a VSA: fence the CLC response read before the slot release; forward PV descriptor lifetime ( #1904 )
...
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com >
2026-10-02 12:37:34 -07:00
8444c0897a
[feat] FastH3 Ref2VA PDD inference with reference-video VSA sparsity ( #1907 )
...
Co-authored-by: Davids048 <jundasu@ucsd.edu >
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com >
2026-10-02 11:28:07 -07:00
Kyle Hu
9491c8638a
[bugfix]: keep the full audio track when saving through the ffmpeg pipe ( #1910 )
2026-10-01 22:22:17 -07:00
Ishan
6809a751fb
[bugfix] Fixed the MiniMax-H3 pin-fallback decode test setup ( #1909 )
2026-10-01 22:17:46 -07:00
Kyle Hu and aryan5v
8322b01815
[bugfix]: restore the NVFP4 module after test_nvfp4_config re-imports it ( #1908 )
...
Co-authored-by: aryan5v <email.aryan1005@gmail.com >
2026-10-01 22:17:32 -07:00
Junda Su
9edc8adf5f
[misc] Move test-suite environment access into the registry and allowlists ( #1898 )
2026-09-30 00:15:39 -07:00
e1f3904799
[kernel] sm_100a CUDA backward for VSA block-sparse attention, 128-token blocks ( #1866 )
...
Co-authored-by: Pengcheng Li <pengchengl@ptyche0203.ptyche.clusters.nvidia.com >
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com >
2026-09-29 16:57:22 -07:00
Junda Su
02f1ce11ae
[misc] Move environment var reads into env registry and rename unprefixed variables ( #1897 )
2026-09-29 14:36:16 -07:00
Aryan Kumar and Aryan Kumar
7f03e03dc6
[docs]: FastH3 V2 cookbook serving and MLX install guide ( #1884 )
...
Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com >
2026-09-29 13:03:18 -07:00
Junda Su
e3b88bb12a
[feat] Add a typed environment-variable registry, policy doc, and contract test ( #1896 )
2026-09-29 12:50:45 -07:00
Junda Su
cb66acd400
[bugfix] Fix environment-variable bugs in LTX-2 debug logging, HF token lookup, and attention backend reads ( #1895 )
2026-09-29 11:48:31 -07:00
li-lizhe
442e2d2e18
[bugfix] fix(metrics): set SceneMetric device_map for any non-CPU device ( #1817 )
2026-09-28 10:24:00 -07:00
Max LI
dd35763ad6
[bugfix] Fix MLX prompt enhancement sampler compatibility ( #1891 )
2026-09-27 14:53:25 -07:00
Keith
e90be598e5
[perf] MiniMax H3: return uint8 frames from the decode worker ( #1828 )
2026-09-24 18:08:56 -07:00
Lele and 武垚乐
ba5e81083c
[bugfix] Preserve video frames when decoded audio is shorter ( #1857 )
...
Signed-off-by: 武垚乐 <wuyaole@mininglamp.com >
Co-authored-by: 武垚乐 <wuyaole@mininglamp.com >
2026-09-24 17:20:37 -07:00
YZJF
76ce9c7fd6
[bugfix] Reject mismatched model names on image generation routes ( #1881 )
2026-09-24 17:19:29 -07:00
Junda Su
08d99c089e
[perf] Reuse cached VAE offload for H3 conditioning ( #1882 )
2026-09-24 17:18:39 -07:00
Jerry-XY
20751a21aa
[bugfix] Keep the VSA-H3 tile buffer out of the autograd graph ( #1858 )
2026-09-24 17:16:49 -07:00
alanhuangyoo and SolitaryThinker
9dd2a837f4
[bugfix] encoders: keep Qwen2.5-VL importable on transformers 5 ( #1791 )
...
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com >
2026-09-24 17:13:23 -07:00
alanhuangyoo
93aab45ac2
[bugfix] tests: include ltx2_3_base in local LTX-2 preset set ( #1750 )
2026-09-24 17:11:31 -07:00
Yogya Mehrotra
017ce6602d
[bugfix] Fix dead NVFP4 QAT plumbing: bench imports, int8 error string, smooth_q crash ( #1723 )
2026-09-24 17:09:33 -07:00
Yogya Mehrotra
a575055eec
[perf] benchmark_attn_qat_train: device peak-TFLOPS table instead of silent RTX 5090 default ( #1724 )
2026-09-24 17:08:03 -07:00
Kyle Hu
81f3fec7fd
[bugfix]: workers killed by a signal now log the reason ( #1725 )
2026-09-24 17:06:55 -07:00
Raghav K
d265a454bf
[bugfix]: Fall back when large pinned output allocation fails ( #1759 )
2026-09-24 17:05:59 -07:00
c100c66578
[bugfix]: isolate cancelled streaming requests from GPU result readers ( #1848 )
...
Co-authored-by: Gxj230958 <222823329+Gxj230958@users.noreply.github.com >
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com >
2026-09-24 17:04:59 -07:00
Aryan Kumar
8760eb7a06
[feat] Add FastH3 8-Step V2 MLX inference ( #1863 )
...
Keep the four-step uniform AdaLN cache, and use explicit DMD rungs only when the snapshot declares its own scheduler shifts.
2026-09-23 12:42:03 -07:00
Aryan Kumar
361f919c88
[feat]: add a narrow FastH3 8-step MLX ladder
...
Keep the four-step uniform AdaLN cache, and use explicit DMD rungs only when the snapshot declares its own scheduler shifts.
2026-09-23 11:06:55 -07:00
Aryan Kumar
8b5377aab2
Merge remote-tracking branch 'origin/main' into aryan/pr-1863-fasth3-8step
2026-09-23 10:52:20 -07:00
Yixing Wang and leo
d995516da0
[bugfix] Restore page interaction after dismissing Create Job or successfully creating a job ( #1869 )
...
Co-authored-by: leo <yixingwang@YIXINGs-MacBook-Pro.local >
2026-09-21 12:03:54 -07:00
Keith
f47ad3f5b7
[bugfix] fastvideo-kernel: install the Python + Triton package when the HIP toolchain cannot be configured ( #1871 )
2026-09-21 11:59:13 -07:00
Hexu Zhao and Claude Opus 5
10bdf5e076
[bugfix] frozen offload: no host copy for an already resident module
...
`load` took the host copies before checking which tensors were missing from
the device, so with `vae_cpu_offload=False` — where the loader builds the VAEs
on CUDA and `unload` is never called — it pulled about 11 GB per rank back over
PCIe once and kept a pinned mirror of it for the life of the process, for a
module that never leaves the device.
It now asks what is absent first and returns early when nothing is. The
offloaded path is unchanged: weights still round-trip and come back bit
identical.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com >
2026-09-20 22:46:13 -07:00
Hexu Zhao and Claude Opus 5
3e26db40b0
[perf] H3 decode: pin the frozen VAEs' host copies
...
With the device-to-host copy gone, the remaining host-to-device copy is the
stage's cost, and from pageable memory it is staged through a bounce buffer at
roughly 2.6 GB/s instead of full PCIe speed. The host copies are made once, so
pinning them costs nothing per request.
It does cost the module's size in non-pageable host memory (10.4 GB for the
H3 video VAE, 0.6 GB for the audio VAE, per rank) for as long as the pipeline
is alive. It follows --pin-cpu-memory, which is on by default; --no-pin-cpu-
memory keeps the copies pageable and only loses the bandwidth. Note that this
flag did not already reach the VAE: its other use is FSDP's own CPUOffloadPolicy
and the VAE does not go through FSDP.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com >
Claude-Session: https://claude.ai/code/session_01ATw7g6cxq9N2LtnMratsru
2026-09-20 22:46:13 -07:00
Hexu Zhao and Claude Opus 5
c73dd0ab55
[perf] H3 decode: offload the frozen VAEs without the device-to-host copy
...
The decode stages move the 10.4 GB fp32 video VAE (and the 0.6 GB audio VAE)
with module.to(device) / module.to("cpu") on every request. The weights are
frozen -- the whole stage runs under no_grad -- so the copy back to the host
returns bytes the host already has.
Keep one host copy per tensor, made the first time the module is loaded, and
on unload point .data back at it instead of copying. One H2D per request, no
D2H at all; the module still lives on the host between requests. Where the old
path allocated a fresh host tensor per cycle and freed the previous one, this
reuses one buffer, so peak host memory can only go down.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com >
Claude-Session: https://claude.ai/code/session_01ATw7g6cxq9N2LtnMratsru
2026-09-20 22:46:13 -07:00
430e52154e
[bugfix]: Restore exact Qwen3-VL vision interpolation ( #1737 )
...
Co-authored-by: William Lin <8941107+SolitaryThinker@users.noreply.github.com >
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com >
2026-09-18 05:00:05 -07:00
vanch
384eee8aef
feat(mlx): add FastH3-8-Step-V2 DMD schedule and thermal-safe MLX inference
...
- Support official 8-step DMD schedule from fastvideo_inference.json in MLX pipeline
- Implement MiniMaxH3SchedulerState.from_dmd_steps for exact sigma ladder calculation
- Add auto-detection of fastvideo_inference.json in checkpoint converter AdaLN precomputation
- Add inter_step_cooldown_s and per-step MLX graph/cache cleanup to prevent thermal throttling
- Support configurable VAE tile sizes to guarantee 256px tiled decode without grid artifacts
- Skip non-existent shard files gracefully in Qwen3-VL conditioner
2026-09-17 16:12:02 +08:00
Keith and SolitaryThinker
c4824c7764
[bugfix] FP8: gate the ROCm _scaled_mm path on CDNA4 (gfx950) instead of the capability tuple ( #1859 )
...
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com >
2026-09-16 19:58:55 -07:00
Raghav K
9b0e57fe4b
[ci] Make Dreamverse provider race test deterministic ( #1729 )
2026-09-15 14:13:56 -07:00
William Lin
0100218594
[feat] Support the FastH3 8-Step V2 checkpoint: checkpoint-defined shifts, explicit DMD schedule, new example ( #1852 )
2026-09-15 14:13:45 -07:00
li-lizhe
39718cd54d
[bugfix] fix(cosmos): make AdaLayerNorm autocast device-agnostic ( #1818 )
2026-09-15 07:44:51 -07:00
Ishan and SolitaryThinker
37d06a832f
[feat] Add fastvideo serve configs for Wan CUDA models ( #1801 )
...
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com >
2026-09-14 18:23:35 -07:00
sudhirpol522
8839ba8d4d
[bugfix] Add OpenAI-compatible image generation endpoint ( #1840 )
2026-09-14 17:55:56 -07:00
