agent evaluation benchmark

• τ²-Bench
  • τ²-Bench Evaluation Benchmark Introduction
  • Quick Start with τ²-Bench Evaluation in AISBench
  • Continue Evaluation After Interruption
  • Multiple Executions of a Single Case (pass^k)
  • Using the TAU2-mini Sampled Subset
• SWEbench User Guide
  • 1. Feature Overview
  • 2. Prerequisites
  • 3. Minimal Configuration (Run First, Tune Later)
  • 4. Run Commands
  • 5. How to Read Outputs
  • 6. Common Issues and Troubleshooting (SWEB Error Codes)
  • 7. Advanced Tips (Optional)
• SWE-Bench Pro User Guide
  • 1. Feature Overview
  • 2. Prerequisites
  • 3. Minimal Configuration (Run First, Tune Later)
  • 4. Run Commands
  • 5. How to Read Outputs
  • 6. Common Issues and Troubleshooting (SWEBP Error Codes)
  • 7. Advanced Tips (Optional)
• Harbor Terminal-Bench
  • Harbor Introduction
  • Quick Start with Harbor Terminal-Bench 2.0 in AISBench
  • Continue Evaluation After Interruption
  • Multiple Executions of a Single Case (pass@k)
  • Task Configuration (in datasets) - Key Parameters
  • Agent Configuration (in models) - Related Parameters