TerminalBench v2.1 benchmark
175 AI models ranked on TerminalBench v2.1. Each row shows the model’s score next to its cheapest API price per 1M tokens where it has one, so you can weigh quality against cost.
Top model today
Claude Fable 5.1 (Anthropic) leads at 91%, from $10.00 per 1M input via Anthropic.
Data via GitHub: harbor-framework/terminal-bench-2-1 →Models ranked
175
on this benchmark
Top score
91%
current leader
Type
Single eval
benchmark
Updated
Oct 14, 2026
benchmark scores
Cost vs TerminalBench v2.1
121 priced models = reasoning model (extended thinking)
Each labelled name links to that model’s page. Curated to the cost-efficiency frontier, top scorers and cheapest so labels stay readable; the full field is in the leaderboard table below (and via “select all”). Reasoning models (extended “thinking” before answering) carry a ring around the dot.
Models ranked on TerminalBench v2.1
highest score first · 175 modelsMaker
| # | ||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Claude Fable 5.1 |  Anthropic | 91% | $10.00 via Anthropic | $50.00 via Anthropic | |||||||||||||||||||||||||||||
| 2 | GPT-5.6 Sol |  OpenAI | 90% | $2.00 Routed quote | $10.00 Routed quote | |||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||
| 3 | Claude Opus 5 | Anthropic | 89% | $5.00 via Anthropic | $25.00 via Anthropic | |||||||||||||||||||||||||||||
| 4 | Qwen3.8 Max (0902) |  Alibaba | 89% | $2.00 Routed quote | $6.00 Routed quote | |||||||||||||||||||||||||||||
| 5 | Grok 4.6 |  xAI | 88% | $2.00 via xAI | $6.00 via xAI | |||||||||||||||||||||||||||||
| 6 | GPT-6 Astra | OpenAI | 88% | $10.00 via OpenAI | $50.00 via OpenAI | |||||||||||||||||||||||||||||
| 7 | GPT-5.6 Terra | OpenAI | 88% | $2.00 via OpenAI | $12.00 via OpenAI | |||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||
| 8 | Gemini 3.8 Flash |  Google | 88% | $0.75 via Google | $3.75 via Google | |||||||||||||||||||||||||||||
| 9 | Qwen3.8-Flash-Next | Alibaba | 86% | n/a | n/a | |||||||||||||||||||||||||||||
| 10 | Gemini 3.7 Flash | Google | 86% | $0.75 via Google | $3.75 via Google | |||||||||||||||||||||||||||||
| 11 | Kimi K3 |  Moonshot | 85% | $0.70 Routed quote | $12.50 via Sail | |||||||||||||||||||||||||||||
| 12 | Claude Fable 5 | Anthropic | 85% | $10.00 via Anthropic | $50.00 via Anthropic | |||||||||||||||||||||||||||||
| 13 | Claude Opus 4.8 | Anthropic | 85% | $5.00 via Anthropic | $25.00 via Anthropic | |||||||||||||||||||||||||||||
| 14 | GLM 5.3 Flash |  Zhipu | 84% | $0.11 via Sail | $0.35 via Sail | |||||||||||||||||||||||||||||
| 15 | GPT-5.5 | OpenAI | 84% | $5.00 via OpenAI | $30.00 via OpenAI | |||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||
| 16 | Muse Spark 1.3 | Other | 84% | $1.25 via Azure | $4.25 via Azure | |||||||||||||||||||||||||||||
| 17 | GLM 5.3 | Zhipu | 84% | $0.079 Routed quote | $3.08 via Sail | |||||||||||||||||||||||||||||
| 18 | Claude Opus 4.7 | Anthropic | 83% | $5.00 via Anthropic | $25.00 via Anthropic | |||||||||||||||||||||||||||||
| 19 | Qwen3.8 2.4T A95B | Alibaba | 82% | $2.00 via Aihubmix | $6.00 via Aihubmix | |||||||||||||||||||||||||||||
| 20 | Grok 4.5 | xAI | 82% | $2.00 via xAI | $6.00 via xAI | |||||||||||||||||||||||||||||
| 21 | Qwen3.8 Max | Alibaba | 81% | $1.65 via DeepInfra | $4.95 via DeepInfra | |||||||||||||||||||||||||||||
| 22 | GPT-5.6 Luna | OpenAI | 81% | $0.20 via OpenAI | $1.20 via OpenAI | |||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||
| 23 | Claude Sonnet 5 | Anthropic | 81% | $2.00 via Anthropic | $10.00 via Anthropic | |||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||
| 24 | Muse Spark 1.2 | Other | 80% | $1.25 via Meta | $4.25 via Meta | |||||||||||||||||||||||||||||
| 25 | Qwen3.8 27B | Alibaba | 80% | $0.04 Routed quote | $1.49 via Cerebras | |||||||||||||||||||||||||||||
| 26 | DeepSeek V4 Flash 0731 |  DeepSeek | 79% | $0.016 Routed quote | $0.153 Routed quote | |||||||||||||||||||||||||||||
| 27 | DeepSeek V4 Pro 0813 | DeepSeek | 79% | $0.66 Routed quote | $1.98 Routed quote | |||||||||||||||||||||||||||||
| 28 | Gemini 3.5 Flash | Google | 79% | $1.50 via Google | $9.00 via Google | |||||||||||||||||||||||||||||
| 29 | GPT-5.4 | OpenAI | 78% | $2.50 via Databricks | $15.00 via OpenAI | |||||||||||||||||||||||||||||
| 30 | GLM 5.2 | Zhipu | 78% | $0.05 Routed quote | $1.98 via Scx AI | |||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||
| 31 | Muse Spark 1.1 | Other | 78% | $1.25 via Meta | $4.25 via Meta | |||||||||||||||||||||||||||||
| 32 | Gemini 3.6 Flash | Google | 78% | $0.75 via Google | $3.75 via Google | |||||||||||||||||||||||||||||
| 33 | Motif 3 | Motif-technologies | 75% | n/a | n/a | |||||||||||||||||||||||||||||
| 34 | Qwen3.7 Max | Alibaba | 75% | $1.25 via Novita | $3.75 via Novita | |||||||||||||||||||||||||||||
| 35 | DeepSeek V4 Flash Vision (Reasoning, Max Effort) | DeepSeek | 74% | n/a | n/a | |||||||||||||||||||||||||||||
| 36 | Gemini 3.1 Pro Preview | Google | 74% | $2.00 via Google | $12.00 via Google | |||||||||||||||||||||||||||||
| 37 | K2 Horizon 375B A23B | Mbzuai-institute-of-foundation-models | 72% | n/a | n/a | |||||||||||||||||||||||||||||
| 38 | Claude Sonnet 4.6 | Anthropic | 71% | $3.00 via Databricks | $15.00 via Anthropic | |||||||||||||||||||||||||||||
| 39 | Motif 3 (Beta) (0714) | Motif-technologies | 71% | n/a | n/a | |||||||||||||||||||||||||||||
| 40 | KAT-Coder-Pro V2 | Other | 70% | $0.30 Routed quote | $1.20 Routed quote | |||||||||||||||||||||||||||||
| 41 | Agnes 2.5 Pro Beta | Sapiens-ai | 70% | n/a | n/a | |||||||||||||||||||||||||||||
| 42 | Apodex 1.1 | Apodex | 70% | n/a | n/a | |||||||||||||||||||||||||||||
| 43 | Quasar 438B | Multiverse-computing | 69% | n/a | n/a | |||||||||||||||||||||||||||||
| 44 | Nex-N2-Pro | Other | 68% | $0.25 Routed quote | $1.00 Routed quote | |||||||||||||||||||||||||||||
| 45 | Kimi K2.7 Code | Moonshot | 67% | $0.671 Routed quote | $3.35 Routed quote | |||||||||||||||||||||||||||||
| 46 | Kimi K2.6 | Moonshot | 66% | $0.65 via Wandb | $3.40 via Novita | |||||||||||||||||||||||||||||
| 47 | MiniMax M3 |  MiniMax | 65% | $0.23 via Wandb | $0.96 via Wandb | |||||||||||||||||||||||||||||
| 48 | MiMo-V2.5-Pro |  Xiaomi | 65% | $0.435 Routed quote | $0.87 Routed quote | |||||||||||||||||||||||||||||
| 49 | DeepSeek V4 Pro 0423 | DeepSeek | 65% | $0.435 via Tencent | $0.87 via Tencent | |||||||||||||||||||||||||||||
| 50 | Hy3 | Tencent | 64% | $0.132 Routed quote | $0.528 Routed quote | |||||||||||||||||||||||||||||
1–50 of 183
Page 1 of 4
On-demand listed rates, OpenRouter routed quotes and Vercel reseller quotes. Enterprise commitments can be materially lower. Cost per task uses best-effort token usage. An OpenRouter routed quote is the price of whichever upstream provider OpenRouter prefers today, so it moves day to day.
Frequently asked questions
What is the TerminalBench v2.1 benchmark?
A verified release of the Terminal-Bench agentic benchmark: 89 curated real-world tasks (model training, sysadmin, data processing, security, software engineering) an agent must complete in a sandboxed container terminal. It refines v2.0 for bugs, timeouts and reward-hacking robustness, scored by automated verification tests.
Which AI model scores highest on TerminalBench v2.1?
Claude Fable 5.1 (Anthropic) leads with 91%, from $10.00 per 1M input tokens from Anthropic.
How many models are ranked on TerminalBench v2.1?
175 models carry a TerminalBench v2.1 score. Scores come from Artificial Analysis; where a model has a listed API price it is drawn from LiteLLM and OpenRouter, and models with no current price are listed without one.
Other benchmarks
compare the same models on a different evalIntelligence Indexcomposite indexCoding Indexcomposite indexAgentic Indexcomposite indexGPQA DiamondleaderboardHumanity's Last ExamleaderboardMMLU-ProleaderboardMMMU-ProleaderboardLiveCodeBenchleaderboardSciCodeleaderboardTerminalBench Hardleaderboardτ²-benchleaderboardτ-bench BankingleaderboardAPEX AgentsleaderboardIT-Bench SREleaderboardGDPvalleaderboardCritPtleaderboardOmniscienceleaderboardIFBenchleaderboardLong-Context Reasoningcomposite indexSWE-bench VerifiedleaderboardSWE-bench ProleaderboardAIME 2026leaderboardARC-AGI-2leaderboardAA-Briefcaseleaderboard
Benchmark scores via Artificial Analysis · scores updated Oct 14, 2026 · Listed prices from LiteLLM, OpenRouter and Vercel · catalogue refreshed Oct 14, 2026 · 80 prices older, back to Jul 19, 2026
Every weekday
AI moves fast. Here's your debrief.
News, analysis, tools, and more.
For people who build with AI