Removed flash-attn as dependency because it's hard to install properly

2025-04-02 14:42:35 +02:00
parent 78c71ab8b4
commit e573b9ebe7
13 changed files with 147 additions and 118 deletions
--- a/sia/llm_engine/qwq_llm_engine.py
+++ b/sia/llm_engine/qwq_llm_engine.py
@@ -37,7 +37,6 @@ class QwQLlmEngine(LlmEngine):
            model_path,
            return_dict=True,
            device_map="auto",
-            attn_implementation="flash_attention_2",
            use_cache=True,
            quantization_config=quantization_config,
        )