Aryan Kumar fed3ce18b6 [fix]: align the QAD configs and docs with the released provenance ...

- V2 config now matches the canonical 16-GPU stage 1: warmup 10,
  min LR ratio 0.3, HSDP 16.
- State that the published V2.1 checkpoint is linears-only (BF16 VSA
  attention); the FP4 adaptation is a separate, non-public stage.
- Name the released checkpoints and starting models.
- Vendor and pin both calibrated amax tables.
- Add the V2 prompt/embedding and OmniRef eval data-prep scripts.

2026-10-19 12:38:43 -07:00

..

datasets

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

2026-08-02 14:21:24 -07:00

distill

[misc]: Refactor separate training pipeline runner (#1696)

2026-10-07 03:19:21 -07:00

inference

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

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

serving

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

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

train

[fix]: align the QAD configs and docs with the released provenance

2026-10-19 12:38:43 -07:00

training

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

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