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 Hard

Source: Artificial Analysis
Version: Not reported
Measurement date: Not reported
Metric: % · Higher is better
Evaluation scope: Hard terminal-agent tasks; observed 2026-10-19; result date not supplied

Exact score ties share a position. Positions do not establish statistical significance. Missing metadata is shown as unreported; matching imported metadata is not independent verification of the protocol.

#1GPT-5.6 Sol
openai/gpt-5.6-sol-20260709
65.9%
View modelEvidence independent
#2Claude Fable 5
anthropic/claude-fable-5
62.9%
View modelEvidence independent
#3GPT-5.6 Terra
openai/gpt-5.6-terra-20260709
57.6%
View modelEvidence independent
#4MiMo-V2.5-Pro
xiaomi/mimo-v2.5-pro-20260422
43.2%
View modelEvidence independent
#5MiniMax M3
minimax/minimax-m3-20260531
42.4%
View modelEvidence independent
#6MiMo-V2.5
xiaomi/mimo-v2.5-20260422
41.7%
View modelEvidence independent
#7Nemotron 3 Ultra
nvidia/nemotron-3-ultra-550b-a55b-20260604
36.4%
View modelEvidence independent
#8Mistral Medium 3
mistralai/mistral-medium-3
33.3%
View modelEvidence independent
#8Mistral Medium 3.1
mistralai/mistral-medium-3.1
33.3%
View modelEvidence independent
#8Mistral Medium 3.5
mistralai/mistral-medium-3.5-20260430
33.3%
View modelEvidence independent
#11MiMo-V2-Flash
xiaomi/mimo-v2-flash
25.8%
View modelEvidence independent
#12Command A
cohere/command-a-03-2025
25%
View modelEvidence independent
#12Command A+
cohere/command-a-plus-05-2026
25%
View modelEvidence independent
#14gpt-oss-120b
openai/gpt-oss-120b
23.5%
View modelEvidence independent
#15Phi 4
microsoft/phi-4
3.8%
View modelEvidence independent