Mungert
/

Qwen2.5-VL-7B-Instruct-GGUF

Image-Text-to-Text

Model card Files Files and versions

Mungert commited on Mar 28

Commit

203cf2c

·

verified ·

1 Parent(s): c7af1a9

Update README.md

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -79,7 +79,7 @@ All tests conducted on **Llama-3-8B-Instruct** using:
 - 2048-token context window
 - Same prompt set across all quantizations
-### **Key Improvements**
 - **Dynamic Precision Allocation**:
   - First/Last 25% of layers → IQ4_XS (selected layers)
   - Middle 50% → IQ2_XXS/IQ3_S (increase efficiency)

 - 2048-token context window
 - Same prompt set across all quantizations
+### **Method**
 - **Dynamic Precision Allocation**:
   - First/Last 25% of layers → IQ4_XS (selected layers)
   - Middle 50% → IQ2_XXS/IQ3_S (increase efficiency)