423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2026-02-27 00:05:58 +08:00
Code Issues Packages Projects Releases Wiki Activity

[fix] fp8 (#9735)

Browse Source 
Co-authored-by: jeremy.zhang <jeremy.zhang@temu.com>
This commit is contained in:
ZZHbible
2026-01-09 13:29:28 +08:00
committed by GitHub
parent 5fb5d7ebd3
commit 5aacbe8434
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
src/llamafactory/hparams/parser.py
View File

@@ -340,7 +340,7 @@ def get_train_args(args: dict[str, Any] | list[str] | None = None) -> _TRAIN_CLS
if training_args.deepspeed is not None and (finetuning_args.use_galore or finetuning_args.use_apollo):
raise ValueError("GaLore and APOLLO are incompatible with DeepSpeed yet.")
if training_args.fp8 and training_args.quantization_bit is not None:
if training_args.fp8 and model_args.quantization_bit is not None:
raise ValueError("FP8 training is not compatible with quantization. Please disable one of them.")
if model_args.infer_backend != EngineName.HF: