Share promptsSign inExplore benchmarks
← Agents & tool use benchmarks

Terminal-Bench 2.1 leaderboard (Sep 2026): AI model scores

As of 28 Sep 2026, the highest reported Google score on Terminal-Bench 2.1 is Gemini 3.8 Flash (Google) at 81.27%, #8 on that leaderboard, per Vals AI. The source no longer updates this leaderboard.

What it measures

A set of difficult terminal-based tasks, now superseded by Terminal-Bench 4.0.

The earlier, widely reported Terminal-Bench task set for command-line agents.

What the score means

Score: tasks where the agent passes all tests

Share of terminal tasks completed and verified by tests.

Results from published AI leaderboards. Compare models within each benchmark; scores from different tests aren’t directly comparable. How we collect these results

Scores by model

Run by Vals AI (benchmark by Terminal-Bench team). Leaderboard dated 28 Sep 2026 (no longer updated by the source). Last checked 22 Oct 2026. Showing top 25 of 76 models listed.

Highlight a provider: Anthropic OpenAI Google xAI Meta DeepSeek Moonshot Xiaomi

Highlighting Google models. Other providers are dimmed. Show all providers

  • Anthropic
  • OpenAI
  • Google
  • xAI
  • Meta
  • DeepSeek
  • Moonshot
  • Xiaomi
Terminal-Bench 2.1: Score vs Cost per test (USD)Scatter plot of 25 models. Leader on Score: Claude Opus 5.5 at 87.64%. Google models highlighted. Full values in the table below.$.01$.1$1$1070%75%80%85%COST PER TEST (USD) · LOG SCALESCOREClaude Opus 4.8 (Anthropic): Cost per test (USD) $2.41, Score 71.91%AMuse Spark 1.3 (Meta): Cost per test (USD) $0.66, Score 72.28%MGPT-6 Luna (OpenAI): Cost per test (USD) $0.030, Score 73.03%OGrok 4.7 (xAI): Cost per test (USD) $1.08, Score 73.41%XClaude Sonnet 5 (Anthropic): Cost per test (USD) $0.53, Score 74.53%ADeepSeek V4.1 Flash (DeepSeek): Cost per test (USD) $0.10, Score 74.53%DMiMo V2.6 Flash (Xiaomi): Cost per test (USD) $0.020, Score 76.40%XiGPT 5.5 (OpenAI): Cost per test (USD) $0.74, Score 76.40%OGPT-5.6 Terra (OpenAI): Cost per test (USD) $0.47, Score 77.53%OGrok 4.6 (xAI): Cost per test (USD) $0.45, Score 78.28%XMuse Spark 1.3 Max (Meta): Cost per test (USD) $0.60, Score 79.03%MGPT-5.6 Luna (OpenAI): Cost per test (USD) $0.050, Score 79.03%OClaude Fable 5 (Anthropic): Cost per test (USD) $1.43, Score 80.52%AKimi K3 (Moonshot): Cost per test (USD) $0.34, Score 80.90%KClaude Sonnet 5.5 (Anthropic): Cost per test (USD) $0.63, Score 83.15%AGPT-6 Sol (OpenAI): Cost per test (USD) $0.37, Score 83.15%OClaude Opus 5 (Anthropic): Cost per test (USD) $0.89, Score 84.64%AClaude Fable 5.1 (Anthropic): Cost per test (USD) $3.07, Score 85.02%AGPT-5.6 Sol (OpenAI): Cost per test (USD) $1.02, Score 85.77%OGPT-6 Astra (OpenAI): Cost per test (USD) $1.34, Score 87.27%OClaude Opus 5.5 (Anthropic): Cost per test (USD) $0.50, Score 87.64%AGemini 3.6 Flash (Google): Cost per test (USD) $1.17, Score 73.78%GGemini 3.5 Flash (Google): Cost per test (USD) $0.87, Score 74.16%GGemini 3.7 Flash (Google): Cost per test (USD) $1.02, Score 77.53%GGemini 3.8 Flash (Google): Cost per test (USD) $1.54, Score 81.27%GGemini 3.8 FlashGemini 3.7 FlashClaude Opus 5.5
Terminal-Bench 2.1: Score vs Cost per test (USD)Scatter plot of 25 models. Leader on Score: Claude Opus 5.5 at 87.64%. Google models highlighted. Full values in the table below.$.01$.1$1$1070%75%80%85%COST PER TEST (USD) · LOG SCALESCOREClaude Opus 4.8 (Anthropic): Cost per test (USD) $2.41, Score 71.91%AMuse Spark 1.3 (Meta): Cost per test (USD) $0.66, Score 72.28%MGPT-6 Luna (OpenAI): Cost per test (USD) $0.030, Score 73.03%OGrok 4.7 (xAI): Cost per test (USD) $1.08, Score 73.41%XClaude Sonnet 5 (Anthropic): Cost per test (USD) $0.53, Score 74.53%ADeepSeek V4.1 Flash (DeepSeek): Cost per test (USD) $0.10, Score 74.53%DMiMo V2.6 Flash (Xiaomi): Cost per test (USD) $0.020, Score 76.40%XiGPT 5.5 (OpenAI): Cost per test (USD) $0.74, Score 76.40%OGPT-5.6 Terra (OpenAI): Cost per test (USD) $0.47, Score 77.53%OGrok 4.6 (xAI): Cost per test (USD) $0.45, Score 78.28%XMuse Spark 1.3 Max (Meta): Cost per test (USD) $0.60, Score 79.03%MGPT-5.6 Luna (OpenAI): Cost per test (USD) $0.050, Score 79.03%OClaude Fable 5 (Anthropic): Cost per test (USD) $1.43, Score 80.52%AKimi K3 (Moonshot): Cost per test (USD) $0.34, Score 80.90%KClaude Sonnet 5.5 (Anthropic): Cost per test (USD) $0.63, Score 83.15%AGPT-6 Sol (OpenAI): Cost per test (USD) $0.37, Score 83.15%OClaude Opus 5 (Anthropic): Cost per test (USD) $0.89, Score 84.64%AClaude Fable 5.1 (Anthropic): Cost per test (USD) $3.07, Score 85.02%AGPT-5.6 Sol (OpenAI): Cost per test (USD) $1.02, Score 85.77%OGPT-6 Astra (OpenAI): Cost per test (USD) $1.34, Score 87.27%OClaude Opus 5.5 (Anthropic): Cost per test (USD) $0.50, Score 87.64%AGemini 3.6 Flash (Google): Cost per test (USD) $1.17, Score 73.78%GGemini 3.5 Flash (Google): Cost per test (USD) $0.87, Score 74.16%GGemini 3.7 Flash (Google): Cost per test (USD) $1.02, Score 77.53%GGemini 3.8 Flash (Google): Cost per test (USD) $1.54, Score 81.27%GGemini 3.8 FlashGemini 3.7 FlashClaude Opus 5.5
Terminal-Bench 2.1: top 15 by ScoreDot plot, axis starting at 65% (not zero). Leader: Claude Opus 5.5 at 87.64%. Google models highlighted. Full values in the table below.65%70%75%80%85%SCORE · AXIS STARTS AT 65%, NOT ZEROClaude Opus 5.5 (Anthropic), rank 1: 87.64%AClaude Opus 5.587.64%GPT-6 Astra (OpenAI), rank 2: 87.27%OGPT-6 Astra87.27%GPT-5.6 Sol (OpenAI), rank 3: 85.77%OGPT-5.6 Sol85.77%Claude Fable 5.1 (Anthropic), rank 4: 85.02%AClaude Fable 5.185.02%Claude Opus 5 (Anthropic), rank 5: 84.64%AClaude Opus 584.64%GPT-6 Sol (OpenAI), rank 6: 83.15%OGPT-6 Sol83.15%Claude Sonnet 5.5 (Anthropic), rank 7: 83.15%AClaude Sonnet 5.583.15%Gemini 3.8 Flash (Google), rank 8: 81.27%GGemini 3.8 Flash81.27%Kimi K3 (Moonshot), rank 9: 80.90%KKimi K380.90%Claude Fable 5 (Anthropic), rank 10: 80.52%AClaude Fable 580.52%GPT-5.6 Luna (OpenAI), rank 11: 79.03%OGPT-5.6 Luna79.03%Muse Spark 1.3 Max (Meta), rank 12: 79.03%MMuse Spark 1.3 Max79.03%Grok 4.6 (xAI), rank 13: 78.28%XGrok 4.678.28%Gemini 3.7 Flash (Google), rank 14: 77.53%GGemini 3.7 Flash77.53%GPT-5.6 Terra (OpenAI), rank 15: 77.53%OGPT-5.6 Terra77.53%Gemini 3.5 Flash (Google), rank 20: 74.16%G#20 Gemini 3.5 Flash74.16%Gemini 3.6 Flash (Google), rank 21: 73.78%G#21 Gemini 3.6 Flash73.78%
Terminal-Bench 2.1: top 12 by ScoreDot plot, axis starting at 60% (not zero). Leader: Claude Opus 5.5 at 87.64%. Google models highlighted. Full values in the table below.60%70%80%AXIS STARTS AT 60%, NOT ZEROClaude Opus 5.5 (Anthropic), rank 1: 87.64%AClaude Opus 5.587.64%GPT-6 Astra (OpenAI), rank 2: 87.27%OGPT-6 Astra87.27%GPT-5.6 Sol (OpenAI), rank 3: 85.77%OGPT-5.6 Sol85.77%Claude Fable 5.1 (Anthropic), rank 4: 85.02%AClaude Fable 5.185.02%Claude Opus 5 (Anthropic), rank 5: 84.64%AClaude Opus 584.64%GPT-6 Sol (OpenAI), rank 6: 83.15%OGPT-6 Sol83.15%Claude Sonnet 5.5 (Anthropic), rank 7: 83.15%AClaude Sonnet 5.583.15%Gemini 3.8 Flash (Google), rank 8: 81.27%GGemini 3.8 Flash81.27%Kimi K3 (Moonshot), rank 9: 80.90%KKimi K380.90%Claude Fable 5 (Anthropic), rank 10: 80.52%AClaude Fable 580.52%GPT-5.6 Luna (OpenAI), rank 11: 79.03%OGPT-5.6 Luna79.03%Muse Spark 1.3 Max (Meta), rank 12: 79.03%MMuse Spark 1.3 M…79.03%Gemini 3.7 Flash (Google), rank 14: 77.53%G#14 Gemini 3.7 F…77.53%Gemini 3.5 Flash (Google), rank 20: 74.16%G#20 Gemini 3.5 F…74.16%Gemini 3.6 Flash (Google), rank 21: 73.78%G#21 Gemini 3.6 F…73.78%
Score
Share of terminal tasks completed and verified by tests.
Cost per test (USD)
Average API cost per task, as published.

All scores

Terminal-Bench 2.1 scores, Google models highlighted
RankModelProviderScoreCost per test (USD)HarnessPrice in / out ($/1M tokens)
1 Claude Opus 5.5Anthropic87.64%$0.50Terminus 2$4 / $20
2 GPT-6 AstraOpenAI87.27%$1.34Terminus 2$10 / $50
3 GPT-5.6 SolOpenAI85.77%$1.02Terminus 2$4 / $20
4 Claude Fable 5.1Anthropic85.02%$3.07Terminus 2$10 / $50
5 Claude Opus 5Anthropic84.64%$0.89Terminus 2$5 / $25
6 GPT-6 SolOpenAI83.15%$0.37Terminus 2$2 / $10
7 Claude Sonnet 5.5Anthropic83.15%$0.63Terminus 2$2 / $10
8 Gemini 3.8 FlashGoogle81.27%$1.54Terminus 2$1.5 / $7.5
9 Kimi K3Moonshot80.90%$0.34Terminus 2$3 / $15
10 Claude Fable 5Anthropic80.52%$1.43Terminus 2$10 / $50
Show 15 more rows
RankModelProviderScoreCost per test (USD)HarnessPrice in / out ($/1M tokens)
11 GPT-5.6 LunaOpenAI79.03%$0.050Terminus 2$0.2 / $1.2
12 Muse Spark 1.3 MaxMeta79.03%$0.60Terminus 2$1.25 / $4.25
13 Grok 4.6xAI78.28%$0.45Terminus 2$2 / $6
14 Gemini 3.7 FlashGoogle77.53%$1.02Terminus 2$1.5 / $7.5
15 GPT-5.6 TerraOpenAI77.53%$0.47Terminus 2$2 / $12
16 GPT 5.5OpenAI76.40%$0.74Terminus 2$5 / $30
17 MiMo V2.6 FlashXiaomi76.40%$0.020Terminus 2$0.14 / $0.28
18 DeepSeek V4.1 FlashDeepSeek74.53%$0.10Terminus 2$0.3 / $1.2
19 Claude Sonnet 5Anthropic74.53%$0.53Terminus 2$2 / $10
20 Gemini 3.5 FlashGoogle74.16%$0.87Terminus 2$1.5 / $9
21 Gemini 3.6 FlashGoogle73.78%$1.17Terminus 2$1.5 / $7.5
22 Grok 4.7xAI73.41%$1.08Terminus 2$2 / $6
23 GPT-6 LunaOpenAI73.03%$0.030Terminus 2$0.1 / $0.5
24 Muse Spark 1.3Meta72.28%$0.66Terminus 2$1.25 / $4.25
25 Claude Opus 4.8Anthropic71.91%$2.41Terminus 2$5 / $25

Source: Vals AI leaderboard ↗ · Terminal-Bench project ↗

Best-ranked Google model here: Gemini 3.8 Flash (Google). Try prompts for Gemini →

Prompts for this kind of work

Copy-ready prompts from our library that match what Terminal-Bench 2.1 tests. Run them on your own material to compare models.

ProductClaudeSetup or integration instructions for Claude CodeGetting a coding agent to connect or deploy something, with only the browser steps left for you.View prompt
ResearchGrok BotMake your bot the operating layer over business softwareI want to progressively eliminate the need for me to operate business software. Your objective is…View prompt Michael | AI + AgentsMichael | AI + Agents @Michael_Fenech_Report or remove
ResearchClaudeSystem prompt for a product agentWriting the instructions for a customer-facing AI agent on a messaging channel.View prompt
ResearchGrok BotConnect your Claude and ChatGPT subscriptions to your botsConnect my Claude and ChatGPT subscriptions so all of my bots can use them. 1. On your computer,…View prompt Corey GanimCorey Ganim @coreyganimReport or remove