Claude Code
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Claude Code MiniMax-M2.7
60Exact Match (EM)
12
Claude Code coding agent prompts
89.1Similarity Score
10
Claude Code 101 CCR instances
96ASR (GPT)
3
Claude Code
43.67C-F Success Rate
3
Claude Code Evaluation Set
26Number of Samples
3
Claude Code
40Successful Tasks Count
3
Claude Code (test)
18.7Total Tokens
2