Usage

Install the terminal-bench CLI

uv tool install terminal-bench

Run the benchmark

tb run \
  --dataset-name terminal-bench-core \
  --dataset-version head \
  --agent terminus \
  --model-name anthropic/claude-3-7-sonnet-latest \
  --livestream \
  --task-id pallets__flask-5014

New to terminal-bench? See our quickstart guide.

Instructions

Require a non-empty name for Blueprints Things do not work correctly if a Blueprint is given an empty name (e.g. #4944). It would be helpful if a `ValueError` was raised when trying to do that.

Tags

debugging, swe-bench