Skip to main content

Benchmark profile

Artificial Analysis Terminal-Bench v2.1 (AA Terminal-Bench 2.1)

An independently evaluated Terminal-Bench v2.1 result from Artificial Analysis.

Data verified

About AA Terminal-Bench 2.1

Year

2026

Tasks

Terminal-based agent tasks

Format

Task success rate

Difficulty

Agentic software engineering

Stored separately from the core Terminal-Bench lane because Artificial Analysis controls the evaluation configuration.

BenchLM freshness & provenance

Version

AA Terminal-Bench 2.1 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

FAQ

What does AA Terminal-Bench 2.1 measure?

An independently evaluated Terminal-Bench v2.1 result from Artificial Analysis.

Which model scores highest on AA Terminal-Bench 2.1?

No models have been evaluated on AA Terminal-Bench 2.1 yet.

How many models are evaluated on AA Terminal-Bench 2.1?

0 AI models have been evaluated on AA Terminal-Bench 2.1 on BenchLM.

Last updated: July 23, 2026 · BenchLM version AA Terminal-Bench 2.1 2026

Choose a model with this week’s evidence

Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.

One email each week. Unsubscribe anytime.