Files

T

Aryan Kumar 69a746c4ce [feat]: Pika H3 Flex NVFP4 QAD profile (Ref2VA + FL2VA, six-forward re-noise ladder)

Add MiniMaxH3FlexRef2VAModel / MiniMaxH3FlexFL2VAModel for OmniRefPDDQADMethod:
dense attention, Pika's shifted 1000..40 ladder, re-noise steps whose noise is a
function of (row seed, forward, modality), NVFP4 linears only. Add an offline
fp32 adapter merge script and QAD configs for both adapters. Let model profiles
pick their attention backend and receive the prepared row in schedulers().

2026-10-29 19:12:47 -07:00

consistency_model/wan

[feat] Add Clean-history Teacher Forcing and Causal Consistency Distillation (#1505 )

2026-07-07 08:13:11 +00:00

distribution_matching

[feat] Add H3 PDD/DMPDD training and packed Ref2VA preprocessing (#1984 )

2026-10-20 11:15:57 -07:00

fine_tuning

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

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

knowledge_distillation

[feat]: Pika H3 Flex NVFP4 QAD profile (Ref2VA + FL2VA, six-forward re-noise ladder)

2026-10-29 19:12:47 -07:00

rl/wan

[feat] Add Wan RL DiffusionNFT training (#1450 )

2026-06-11 21:18:59 -07:00

example.yaml

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

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

overfit_cosmos25_t2w.yaml

[feat] Cosmos 2.5 training support in fastvideo.train (#1224 )

2026-05-01 01:15:02 +00:00

overfit_hunyuan15_t2v.yaml

[feat] Add HunyuanVideo 1.5 T2V training support (#1662 )

2026-10-07 02:19:22 -07:00

overfit_hunyuan_t2v.yaml

[bugfix]: fix VAE temporal tiling blend corruption in tiled_encode (#1181 )

2026-03-29 23:12:42 +00:00

overfit_ltx2_3_t2v.yaml

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

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

overfit_ltx2_t2v_nvfp4_qat.yaml

[feat]: FA4-FP4 ATTN_QAT_INFER on sm_100/sm_103 + NVFP4 weight purge (#1647 )

2026-07-27 15:46:52 -07:00

overfit_ltx2_t2v.yaml

[feat] Add LTX-2 and LTX-2.3 fine-tuning to the modular trainer (#1624 )

2026-07-20 11:58:39 -07:00

overfit_minimax_h3_ref2va_lora.yaml

[feat]: add MiniMax H3 Ref2VA and LoRA training support (#1757 )

2026-10-07 03:21:45 -07:00

overfit_minimax_h3_ref2va.yaml

[feat]: add MiniMax H3 Ref2VA and LoRA training support (#1757 )

2026-10-07 03:21:45 -07:00

overfit_minimax_h3_t2va_lora.yaml

[feat]: add MiniMax H3 Ref2VA and LoRA training support (#1757 )

2026-10-07 03:21:45 -07:00

overfit_minimax_h3_t2va.yaml

[feat] add Minimax H3 sft pipeline (#1688 )

2026-08-07 16:12:04 -07:00

README.md

[misc]: reorganize training configs and add documentation (#1177 )

2026-03-26 16:38:31 -07:00

README.md

Training Configs

Single-step training configurations organized by method and model.

configs/
├── fine_tuning/          # Standard finetuning and DFSFT
├── distribution_matching/  # DMD2 and Self-Forcing
├── knowledge_distillation/ # KD from teacher to student
└── example.yaml          # Annotated reference config with all fields

Each method directory contains per-model subdirectories (e.g. wan/, hunyuan/).

Launch any config with:

bash examples/train/run.sh examples/train/configs/<method>/<model>/<config>.yaml

For multi-step training pipelines, see examples/train/scenario/.