Model results

Terminal-Bench

Choose a shared evaluation protocol to compare model scores. Versions, sources and configurations stay separate.

Atlas snapshot captured 10/19/2026, 10:20:28 AM UTC. This is an import timestamp, not the evaluation date. Coverage may be incomplete.

Reviewed recommendation and limits

Start with Terminal-Bench 4.0 for terminal-based agent work. Compare the same task release, agent and resource settings.

Terminal-Bench 4.0.0 on Harbor. Preserve agent, model, reasoning effort, task resources, trial count and resolution-rate uncertainty. Earlier 2.0, 2.1 and 3.0 results remain separate.. Imported groups below may use different protocols; this review does not certify each run. Public solutions, reward hacking and task-specific tuning can inflate results. Earlier editions and vendor-reported runs do not establish the current reviewed protocol. Terminal completion is different from repository issue repair.

Publisher results Terminal-Bench 4.0 Leaderboard integrity policy Contributors

← All evaluation groups

Terminal-Bench 2.0

Source: Terminal-Bench via Epoch AI
Version: 2.0
Measurement date: Not reported
Metric: % · Higher is better
Evaluation scope: Agent OpenHands · result date not supplied

Positions are withheld because at least one result lacks explicit independent provenance or a documented evaluation scope. Missing metadata is shown as unreported; matching imported metadata is not independent verification of the protocol.

Claude Sonnet 4.5
anthropic/claude-sonnet-4-5
42.6%
View modelEvidence aggregated
Claude Opus 4.1
anthropic/claude-opus-4-1
36.9%
View modelEvidence aggregated
Kimi K2 0711
moonshotai/kimi-k2
26.7%
View modelEvidence aggregated
Qwen3 Coder 480B A35B
qwen/qwen3-coder-480b-a35b-07-25
25.4%
View modelEvidence aggregated
Gemini 2.5 Flash
google/gemini-2.5-flash
16.4%
View modelEvidence aggregated
Gemini 2.5 Pro
google/gemini-2.5-pro
16.4%
View modelEvidence aggregated
Claude Haiku 4.5
anthropic/claude-haiku-4-5
13.9%
View modelEvidence aggregated