423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2025-12-24 15:50:35 +08:00
Code Issues Packages Projects Releases Wiki Activity

[data] fix template (#8827)

Browse Source
This commit is contained in:
Yaowei Zheng
2025-08-06 06:58:09 +08:00
committed by GitHub
parent 4dfad24902
commit b523543994
4 changed files with 8 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
src/llamafactory/model/model_utils/quantization.py
View File

@@ -90,12 +90,13 @@ def configure_quantization(
if model_args.quantization_bit is not None:
logger.warning_rank0("`quantization_bit` will not affect on the PTQ-quantized models.")
if is_deepspeed_zero3_enabled() or is_fsdp_enabled():
raise ValueError("DeepSpeed ZeRO-3 or FSDP is incompatible with PTQ-quantized models.")
quantization_config: dict[str, Any] = getattr(config, "quantization_config", None)
quant_method = quantization_config.get("quant_method", "")
if quant_method != QuantizationMethod.MXFP4 and (is_deepspeed_zero3_enabled() or is_fsdp_enabled()):
# mxfp4 will dequant the model weights
raise ValueError("DeepSpeed ZeRO-3 or FSDP is incompatible with PTQ-quantized models.")
if quant_method == QuantizationMethod.GPTQ:
check_version("gptqmodel>=2.0.0", mandatory=True)
quantization_config.pop("disable_exllama", None) # remove deprecated args