Aryan Kumar cc9bfd069d [perf]: Load H3 DiT weights onto CUDA, not CPU ...

GB10 unified memory made the CPU-then-copy path take minutes for
the 35B FastH3 transformer. When host offload is off, read
safetensors onto the accelerator.

2026-08-30 23:27:24 -07:00

..

api

[perf]: load pipeline components on demand and free them after their last stage

2026-08-28 22:36:51 -07:00

attention

[bugfix]: share compiled VSA graphs across H3 layers

2026-08-29 16:37:51 -07:00

benchmarks

[feat]: Apple Silicon MLX runtime — INT8 Wan2.1 and Wan2.2 inference (#1638)

2026-08-18 15:08:00 -07:00

configs

[perf] Optimize MiniMax-H3 text encoder memory (#1732)

2026-08-21 14:57:06 -07:00

dataset

[misc] yapf pass under CI's interpreter (3.12) + pin hook language_version (#1702)

2026-08-08 22:24:17 -07:00

distributed

[perf] Add fused NVLink all-to-all for Ulysses (#1740)

2026-08-22 23:06:24 -07:00

entrypoints

[bugfix] Harden OpenAI serving after post-merge review (#1782)

2026-08-28 22:29:24 -07:00

eval

[feat] World model training using third person games (#1443)

2026-07-10 05:09:07 +00:00

hooks

[perf] Mark LayerwiseOffloadHook entry points torch.compiler.disable (remove per-layer graph break) (#1365)

2026-05-20 03:09:23 +00:00

layers

[feat] Add MiniMax H3 LoRA inference and preview launchers (#1771)

2026-08-27 14:43:39 -07:00

logging_utils

[Feature] Remove V1 folder (#642)

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

mlx_runtime

[feat] Add MiniMax H3 MLX T2VA inference (#1770)

2026-08-28 12:54:33 -07:00

models

[perf]: Load H3 DiT weights onto CUDA, not CPU

2026-08-30 23:27:24 -07:00

performance

[ci] Add exact identity performance statuses (#1560)

2026-07-15 14:59:09 -07:00

performance_dashboard

[ci] Add exact identity performance statuses (#1560)

2026-07-15 14:59:09 -07:00

pipelines

[bugfix]: keep unused LoRA setup from loading lazy DiTs

2026-08-28 22:36:51 -07:00

platforms

[perf]: MiniMax H3 on GB10 - skip text encoder CPU offload on unified memory (5m49s to 30ms) (#1710)

2026-08-26 12:03:14 -07:00

tests

[bugfix]: share compiled VSA graphs across H3 layers

2026-08-29 16:37:51 -07:00

third_party

[feat]: Apple Silicon MLX runtime — INT8 Wan2.1 and Wan2.2 inference (#1638)

2026-08-18 15:08:00 -07:00

train

[ci] make GPU validation change-aware (#1747)

2026-08-25 21:26:25 -07:00

training

[ci] make GPU validation change-aware (#1747)

2026-08-25 21:26:25 -07:00

worker

[bugfix] Harden OpenAI serving after post-merge review (#1782)

2026-08-28 22:29:24 -07:00

workflow

[misc]: consolidate dataset download scripts under examples/datasets/ (#1667)

2026-08-02 14:21:24 -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] add hierarchical AGENTS.md per-directory guidance (#1278)

2026-05-02 03:28:18 -07:00

envs.py

[perf] Add fused NVLink all-to-all for Ulysses (#1740)

2026-08-22 23:06:24 -07:00

fastvideo_args.py

[perf]: load pipeline components on demand and free them after their last stage

2026-08-28 22:36:51 -07:00

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

[perf] Keep odd MiniMax-H3 VSA tiles on sm100a (#1745)

2026-08-22 18:29:30 -07:00

profiler.py

[perf] Optimize MiniMax H3 VAE decoding (#1734)

2026-08-21 14:57:32 -07:00

registry.py

[feat] Align multimodal OpenAI serving APIs (#1781)

2026-08-28 10:09:58 -07:00

utils.py

[misc]: re-run yapf on main so pre-commit passes again (#1700)

2026-08-07 16:17:52 -07:00

version.py

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

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