Download data/processed/RQ1/EmailResponder-MCP/evaluation_results.csv from AINativeOps/AINativeBench: direct link, hf CLI and curl.
- Browser
- Download file 816 Bytes
-
https://huggingface.co/datasets/AINativeOps/AINativeBench/resolve/main/data/processed/RQ1/EmailResponder-MCP/evaluation_results.csv
- Command line
-
hf download hf://datasets/AINativeOps/AINativeBench/data/processed/RQ1/EmailResponder-MCP/evaluation_results.csv
-
curl -L -o evaluation_results.csv https://huggingface.co/datasets/AINativeOps/AINativeBench/resolve/main/data/processed/RQ1/EmailResponder-MCP/evaluation_results.csv
816 Bytes
| model,num_samples,exact_match,in_order_match,any_order_match,precision,recall,single_tool_use,unique_path_ratio,path_entropy | |
| GPT-5,100,0.0,0.0,0.0,1.0,0.8333333333333335,0.33333333333333326,0.01,0.0 | |
| GPT-4o-mini,100,0.56,0.61,0.61,0.9902857142857142,0.935,0.5366666666666666,0.1,0.7563945757337693 | |
| DeepSeek-V3-1,100,0.06,0.22,0.22,0.9658928571428571,0.8766666666666667,0.4066666666666666,0.12,0.657451991546922 | |
| DeepSeek-R1,100,0.0,0.03,0.03,0.8023767068048835,0.895,0.34,0.3,0.78748356089041 | |
| Gemini-2.5-flash,100,0.0,0.0,0.0,0.9746107250592544,0.7533333333333333,0.12,0.16,0.5735990446233518 | |
| Gemini-2.5-flash-nothinking,100,0.02,0.04,0.05,0.9647254901960785,0.8041666666666667,0.2533333333333333,0.14,0.564918773302568 | |
| Qwen3-235b,100,0.04,0.05,0.05,0.9960714285714286,0.8441666666666667,0.35,0.08,0.35307060153243225 | |