Update README.md

README.md

@@ -20,4 +20,17 @@ The only difference from Llama-3.2-1B-chatml-tool-v1 is that it uses AlternateTo
 
 In the case of the existing tool-AlternateTokenizer, the <tool_call> tag was not properly generated before the function call, but in v2, it was observed that it performed well when trained with the general AlternateTokenizer.
 
-need to check whether this phenomenon is repeated in larger models (3B, 8B).
+need to check whether this phenomenon is repeated in larger models (3B, 8B).
+
+
+
+## Model Performance Comparison (BFCL)
+
+| task name        | minpeter/Llama-3.2-1B-chatml-tool-v2 | meta-llama/Llama-3.2-1B-Instruct |
+|-----------------|-----------------------------------|-----------------------------------|
+| parallel_multiple | 0.000                               | 0.025                               |
+| parallel        | 0.000                               | 0.035                               |
+| simple          | 0.72                             | 0.215                               |
+| multiple        | 0.695                                | 0.17                                |
+
+*Parallel calls are not taken into account. 0 points are expected. We plan to fix this in v3.