Back to Benchmarks
💻
Terminal-Bench 2.0
Terminal Benchmark v2
Command-line and DevOps task completion
💻 Coding
Best: 90.2%
14 Models Evaluated
Champion
GPT-5.5
OpenAI
75.1%
Best Score
Model Rankings
All models ranked by score on Terminal-Bench 2.0. Scores are relative to the best score of 90.2%.
| # | Model | Provider | Score | vs Best |
|---|---|---|---|---|
| 1 | GPT-5.5 Elo 1478 | OpenAI | 75.1% | Best |
| 2 | GPT-5.4 Elo 1463 | OpenAI | 75.1% | -15.1% |
| 3 | Claude Opus 4.6 Elo 1503 | Anthropic | 65.4% | -24.8% |
| 4 | Claude Sonnet 4.6 Elo 1460 | Anthropic | 59.1% | -31.1% |
| 5 | Gemini 3.1 Pro Elo 1492 | 56.2% | -34.0% | |
| 6 | GLM-5 Elo 1454 | Zhipu AI | 56.2% | -34.0% |
| 7 | Qwen 3.5 Elo 1450 | Alibaba | 52.5% | -37.7% |
| 8 | Grok 3 Elo 1412 | xAI | 52% | -38.2% |
| 9 | Step-3.5-Flash Elo 1389 | Stepfun | 51% | -39.2% |
| 10 | Kimi K2.5 Elo 1438 | Moonshot | 50.8% | -39.4% |
| 11 | MiniMax M2.5 Elo 1404 | MiniMax | 42.2% | -48.0% |
| 12 | DeepSeek V3.2 Elo 1423 | DeepSeek | 39.6% | -50.6% |
| 13 | MiMo-V2-Flash Elo 1393 | Xiaomi | 38.5% | -51.7% |
| 14 | GPT-oss 120B Elo 1355 | OpenAI | 18.7% | -71.5% |
Score Distribution
GPT-5.5
75.1%
GPT-5.4
75.1%
Claude Opus 4.6
65.4%
Claude Sonnet 4.6
59.1%
Gemini 3.1 Pro
56.2%
GLM-5
56.2%
Qwen 3.5
52.5%
Grok 3
52%
Step-3.5-Flash
51%
Kimi K2.5
50.8%
Run This Benchmark
Learn how to evaluate your own models on Terminal-Bench 2.0.
Terminal-Bench 2.0 is a standardized benchmark for evaluating AI model capabilities.
To run this benchmark on your own models, visit the official benchmark page or leaderboard for detailed instructions, code, and evaluation scripts.
Data compiled from official model cards, public leaderboards, and independent evaluations.