fix #3365

2024-04-21 19:20:18 +08:00 · 2024-04-21 19:20:18 +08:00 · a1d31ffc8c
parent 5c62881c5a
commit a1d31ffc8c
1 changed files with 1 additions and 0 deletions
--- a/src/llmtuner/model/patcher.py
+++ b/src/llmtuner/model/patcher.py
@ -139,6 +139,7 @@ def _configure_quantization(

        if quant_method == QuantizationMethod.GPTQ:
            require_version("auto_gptq>=0.5.0", "To fix: pip install auto_gptq>=0.5.0")
+            quantization_config.pop("disable_exllama", None)  # remove deprecated args
            quantization_config["use_exllama"] = False  # disable exllama

        if quant_method == QuantizationMethod.AWQ: