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 django__django-14434

New to terminal-bench? See our quickstart guide.

Instructions

Statement created by _create_unique_sql makes references_column always false Description This is due to an instance of Table is passed as an argument to Columns when a string is expected.

Tags

debugging, swe-bench