# Terminal-Bench

Terminal-Bench is an AI benchmark; this page records what it measures and traces its data, grader, and scores to primary sources.

Tests whether agents complete end-to-end tasks inside reproducible terminal environments.

## Details

- Address: Online, Global
- Website: [github.com/harbor-framework/terminal-bench-1](<https://github.com/harbor-framework/terminal-bench-1>)
- Tags: Helpful, Viral, Science
- Category: Agents
- Status: Active
- Data licence: Apache-2.0
- Data access: open
- Grader: unit-test
- Organization: Harbor Framework
- Released: 2025-01
- Leaderboard: [www.tbench.ai/](<https://www.tbench.ai/>)
- Repository: [github.com/harbor-framework/terminal-bench-1](<https://github.com/harbor-framework/terminal-bench-1>)
- Task count: 100
- Top-score provenance: unknown
- ABI family: No

## Questions

### Where is Terminal-Bench?

Terminal-Bench is at Online, Global.

### How do I contact Terminal-Bench?

To contact Terminal-Bench, visit https://github.com/harbor-framework/terminal-bench-1.

## Source

From [Benchmark primary sources](https://aibenchmarkindex.com/about), license CC0-1.0, checked 2026-10-15.

Last updated October 15, 2026.

## Links

- This benchmark on the [web page](<https://aibenchmarkindex.com/global/online/terminal-bench>)
- [Agents in Online](<https://aibenchmarkindex.com/global/online/category/agents.md>)
- [More benchmarks in Online](<https://aibenchmarkindex.com/global/online.md>)

> **Check before you compare.** Benchmark methods and leaderboards change. Follow the cited primary source before using a result in a decision.