Wing Lian 28283ff373 revert shared_kv_states workaround with transformers 5.5.4

2026-04-15 13:32:59 +00:00

..

test_gemma4_fused_attn.py

revert shared_kv_states workaround with transformers 5.5.4

2026-04-15 13:32:59 +00:00

test_gemma4_hybrid_mask.py

[gemma4] fix fused RMSNorm+RoPE on hybrid attention models

2026-04-15 13:27:31 +00:00

test_llama_attn_hijack_flash.py

adding pre-commit auto-update GH action and bumping plugin versions (#2428)

2025-03-21 11:02:43 -04:00

test_pixtral_flash_attention_patch.py

Feat: add Magistral Small 2509 and native mistral3 tokenizer support (#3165)

2025-09-18 15:42:20 +07:00

test_qwen3_next_modeling_patch.py

Feat: add qwen3-next (w packing+cce) (#3150)

2025-09-23 11:31:15 +07:00

test_trainer_accelerator_args.py

basic torchao fp8 mixed precision training (#2926)

2025-07-22 16:27:47 -04:00

test_trainer_context_parallel_patch.py

Cp fix (#3182)

2025-09-25 12:03:50 -04:00

test_trl_vllm.py

Support for Async GRPO (#3486)

2026-03-17 11:42:47 -04:00

test_voxtral_modeling_patch.py

Feat: add Magistral Small 2509 and native mistral3 tokenizer support (#3165)

2025-09-18 15:42:20 +07:00