Kyle Hu ce970eb715 [perf]: training-free token reduction for MiniMax H3 (#2004)

2026-11-08 04:06:38 +00:00

..

api

[feat] Per-request community LoRA in the OpenAI video API (#2026)

2026-11-04 19:44:47 -08:00

attention

[bugfix] Gate the FA4 NVFP4 path on the installed kernel, not the import (#2067)

2026-11-04 19:50:30 -08:00

benchmarks

[misc] Move environment var reads into env registry and rename unprefixed variables (#1897)

2026-09-29 14:36:16 -07:00

configs

[perf]: training-free token reduction for MiniMax H3 (#2004)

2026-11-08 04:06:38 +00:00

dataset

[bugfix] Stop the preprocessing dataset from keeping every decoded video in memory (#2069)

2026-11-06 02:31:54 +00:00

distributed

[bugfix] SP warmup: use the rank's own device instead of hardcoded CUDA (#1988)

2026-11-04 20:28:49 +00:00

entrypoints

[bugfix] Streaming: answer malformed client frames with invalid_message instead of dropping the session (#2073)

2026-11-05 18:14:53 +00:00

eval

[bugfix] Repair DreamVerse and failing CI gates (#1958)

2026-10-16 17:15:38 -07:00

hooks

[feat]: FastH3 on single RTX GPUs (#1919)

2026-10-06 11:37:01 -07:00

layers

[feat] Per-request community LoRA in the OpenAI video API (#2026)

2026-11-04 19:44:47 -08:00

logging_utils

[Feature] Remove V1 folder (#642)

2025-07-24 22:43:12 -07:00

mlx_runtime

[perf] Cache MLX H3 text refinement and speed up attention (#2002)

2026-10-23 12:40:47 -07:00

models

[perf]: training-free token reduction for MiniMax H3 (#2004)

2026-11-08 04:06:38 +00:00

performance

[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

performance_dashboard

[ci] Clarify dashboard cohort grouping for legacy v1 records (#1613)

2026-10-07 02:33:18 -07:00

pipelines

[perf]: training-free token reduction for MiniMax H3 (#2004)

2026-11-08 04:06:38 +00:00

platforms

[feat] ROCm: route the video sparse attention backends to their Triton kernels (#1851)

2026-10-07 02:32:27 -07:00

tests

[perf]: training-free token reduction for MiniMax H3 (#2004)

2026-11-08 04:06:38 +00:00

third_party

[misc] Remove DreamX World, HunyuanGameCraft, GLM-Image, LongCat, TurboDiffusion, and Z-Image (#1926)

2026-10-14 23:16:32 -07:00

train

[bugfix] Map merged-LoRA export keys back to HF names when LoRA was enabled after loading (#2070)

2026-11-05 15:40:06 +00:00

training

[refactor] VSA: rename the sm100a block-sparse kernel to ptx_blackwell (#1956)

2026-10-17 21:10:19 -07:00

worker

[feat] Per-request community LoRA in the OpenAI video API (#2026)

2026-11-04 19:44:47 -08:00

workflow

[feat] Add end-to-end MMAudio V2A training, inference, and evaluation (#1648)

2026-10-07 04:25:00 -07:00

__init__.py

[feat] [4/n] Improve API: refactor sampling param and merge with presets (#1234)

2026-04-16 14:10:02 -07:00

AGENTS.md

[docs]: refresh AGENTS.md maps and add Wan SP/I2V and CI test notes (#1845)

2026-09-14 16:06:53 -07:00

envs.py

[perf] Fused Q/K preconditioning + NVFP4 quantize for FA4 NVFP4 attention (bit-identical) (#2052)

2026-11-03 01:07:11 -08:00

fastvideo_args.py

[perf]: training-free token reduction for MiniMax H3 (#2004)

2026-11-08 04:06:38 +00:00

forward_context.py

[feat] pre-commit support 120 col num (#1167)

2026-03-11 20:19:30 -07:00

image_processor.py

[feat] pre-commit support 120 col num (#1167)

2026-03-11 20:19:30 -07:00

logger.py

[misc]: add env var to control process-aware logging (#1680)

2026-10-07 04:31:52 -07:00

profiler.py

[feat] Add a typed environment-variable registry, policy doc, and contract test (#1896)

2026-09-29 12:50:45 -07:00

registry.py

[new-model] Open-H3-Regenerate-2K: 768P H3 draft to 2560x1440, served on the OpenAI video routes (#2035)

2026-11-01 22:45:09 -08:00

utils.py

[feat] Add end-to-end MMAudio V2A training, inference, and evaluation (#1648)

2026-10-07 04:25:00 -07:00

version.py

[chore]: release v0.2.1 (#1778)

2026-08-28 02:03:18 -07:00