BlogLLM Benchmark Scores as Published, Side by Side
Terminal-Bench: Results as Published
Every result the catalog holds for Terminal-Bench: 123 models from 15 developers, read from 80 pages. Each number is as printed by the company named beside it; nothing is ranked or rescored. Models are listed by developer, newest first.
Area Agentic CodingMaintainer Stanford University and Laude InstituteIntroduced 2025Data version 2026103001, as of 2026-10-30
Terminal-Bench tests whether an AI agent can finish real jobs by typing commands into a Linux terminal, such as building software, configuring a system or processing files. Each task runs in its own sandboxed container, and the agent succeeds only if the final state of that container passes the task's automatic checks. Version 2.0 (November 2025) has 89 hand-checked tasks, and the companion Harbor framework runs the agents.
Terminal-Bench 2.1
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| GPT-5.6 LunaOpenAI, released 2026-07-09 | 84.7% | OpenAI, GPT-5.6: Frontier intelligence that scales with your ambitionchecked 2026-10-15; printed on 2 more pages | - |
| GPT-5.6 SolOpenAI, released 2026-07-09 | 91.9% | OpenAI, GPT-5.6: Frontier intelligence that scales with your ambitionchecked 2026-10-15; printed on 7 more pages | Effort: ultra |
| 88.8% | OpenAI, GPT-5.6: Frontier intelligence that scales with your ambitionchecked 2026-10-15 | - | |
| GPT-5.6 TerraOpenAI, released 2026-07-09 | 87.4% | OpenAI, GPT-5.6: Frontier intelligence that scales with your ambitionchecked 2026-10-15; printed on 4 more pages | - |
| GPT-5.5OpenAI, released 2026-04-23 | 85.6% | OpenAI, GPT-5.6: Frontier intelligence that scales with your ambitionchecked 2026-10-15; printed on 7 more pages | - |
| GPT-5.4 miniOpenAI, released 2026-03-17 | 59.2% | Google, Gemini 3.5 Flash-Lite model cardchecked 2026-10-15 | Harness: Terminus-2 |
| gpt-oss-20bOpenAI, released 2025-08-05 | 15.17% | NVIDIA, NVIDIA Nemotron 3.5 Lightning 30B-A3B model cardchecked 2026-10-19 | - |
| Claude Opus 5Anthropic, released 2026-07-24 | 89.1% | DeepSeek, DeepSeek-V4.1-Flash model cardchecked 2026-10-15; printed on 2 more pages | - |
| Claude Sonnet 5Anthropic, released 2026-06-30 | 80.4% | Anthropic, Claude Sonnet 5checked 2026-10-15; printed on 4 more pages | - |
| Claude Fable 5Anthropic, released 2026-06-09 | 84.6% | Alibaba, Qwen3.8-Max comparisonchecked 2026-10-14; printed on 7 more pages | - |
| Claude Mythos 5Anthropic, released 2026-06-09 | 88% | Anthropic, Claude Fable 5 and Claude Mythos 5checked 2026-10-15; printed on 1 more page | - |
| Claude Opus 4.8Anthropic, released 2026-05-28 | 74.6% | Anthropic, Claude Opus 4.8checked 2026-10-15; printed on 11 more pages | Harness: Terminus-2 |
| Claude Opus 4.7Anthropic, released 2026-04-16 | 66.1% | Anthropic, Claude Opus 4.8checked 2026-10-15; printed on 3 more pages | Harness: Terminus-2 |
| Claude Sonnet 4.6Anthropic, released 2026-02-17 | 67% | Anthropic, Claude Sonnet 5checked 2026-10-15 | - |
| Claude Opus 4.6Anthropic, released 2026-02-05 | 78.2% | Alibaba, Qwen3.8-27B model cardchecked 2026-10-15 | Effort: max; Harness: Terminus |
| Claude Haiku 4.5Anthropic, released 2025-10-15 | 44.2% | Google, Gemini 3.5 Flash-Lite model cardchecked 2026-10-15 | Harness: Terminus-2 |
| Gemini 3.8 FlashGoogle, released 2026-09-02 | 89.4% | Google, Gemini 3.8 Flash model cardchecked 2026-10-15 | - |
| Gemini 3.7 FlashGoogle, released 2026-08-13 | 85.8% | Google, Gemini 3.7 Flash model cardchecked 2026-10-15; printed on 2 more pages | - |
| Gemini 3.5 Flash-LiteGoogle, released 2026-07-21 | 54% | Google, Gemini 3.5 Flash-Lite model cardchecked 2026-10-15 | Harness: Terminus-2 |
| Gemini 3.6 FlashGoogle, released 2026-07-21 | 78% | Google, Gemini 3.6 Flash model cardchecked 2026-10-15; printed on 1 more page | Harness: Terminus-2 |
| Gemini 3.5 FlashGoogle, released 2026-05-19 | 76.2% | Google, Gemini 3.5 Flash model cardchecked 2026-10-15; printed on 1 more page | Harness: Terminus-2 |
| Gemma 4 26B A4BGoogle, released 2026-04-02 | 37.22% | NVIDIA, NVIDIA Nemotron 3.5 Bolt 74B-A9B model cardchecked 2026-10-19; printed on 1 more page | - |
| Gemini 3.1 Flash-LiteGoogle, released 2026-03-03 | 31% | Google, Gemini 3.5 Flash-Lite model cardchecked 2026-10-15 | Harness: Terminus-2 |
| Gemini 3.1 Pro PreviewGoogle, released 2026-02-19 | 70.3% | Google, Gemini 3.5 Flash model cardchecked 2026-10-15; printed on 7 more pages | Harness: Terminus-2 |
| Gemini 3 FlashGoogle, released 2025-12-17 | 58% | Google, Gemini 3.5 Flash model cardchecked 2026-10-15 | Harness: Terminus-2 |
| Grok 4.5xAI, released 2026-07-16 | 83.3% | xAI, Grok 4.5checked 2026-10-15; printed on 1 more page | - |
| Muse Spark OpenMeta, released 2026-10-15 | 76.4% | Meta, Meta Muse Spark Open 816B-A42B model cardchecked 2026-10-19 | Effort: xhigh; Harness: Terminus 2 |
| Muse Glimmer-30BMeta, released 2026-08-09 | 51.7% | Alibaba, Qwen3.8-27B model cardchecked 2026-10-15 | Harness: Terminus |
| Muse Spark 1.2Meta, released 2026-08-05 | 82.9% | Google, Gemini 3.7 Flash model cardchecked 2026-10-15 | - |
| DeepSeek-V4.1-FlashDeepSeek, released 2026-09-10 | 84.1% | DeepSeek, DeepSeek-V4.1-Flash model cardchecked 2026-10-15 | Effort: max; Harness: Codex; Attempt budget: N=3 samples per task; Context: 1M |
| 90.3% | DeepSeek, DeepSeek-V4.1-Flash model cardchecked 2026-10-15 | Effort: max; Harness: mini-SWE; Attempt budget: N=3 samples per task; Context: 1M | |
| 85.8% | DeepSeek, DeepSeek-V4.1-Flash model cardchecked 2026-10-15 | Effort: max; Harness: DeepSeek Harness (DSH) Standard; Attempt budget: N=3 samples per task; Context: 1M | |
| 90.6% | DeepSeek, DeepSeek-V4.1-Flash model cardchecked 2026-10-15 | Effort: max; Harness: DeepSeek Harness (DSH) Minimal; Attempt budget: N=3 samples per task (scaffold table); Context: 1M | |
| 85% | DeepSeek, DeepSeek-V4.1-Flash model cardchecked 2026-10-15 | Effort: max; Harness: OpenCode; Attempt budget: N=3 samples per task; Context: 1M | |
| 88% | DeepSeek, DeepSeek-V4.1-Flash model cardchecked 2026-10-15 | Effort: max; Harness: Claude Code; Attempt budget: N=3 samples per task; Context: 1M | |
| 86.1% | DeepSeek, DeepSeek-V4.1-Flash model cardchecked 2026-10-15 | Effort: max; Harness: Pi; Attempt budget: N=3 samples per task; Context: 1M | |
| 85.8% | DeepSeek, DeepSeek-V4.1-Flash model cardchecked 2026-10-15 | Effort: max; Harness: DeepSeek Harness (DSH) PTC; Attempt budget: N=3 samples per task; Context: 1M | |
| DeepSeek-V4-Flash (Preview)DeepSeek, released 2026-04-24 | 61.8% | DeepSeek, DeepSeek-V4-Flash-0731 model cardchecked 2026-10-15; printed on 2 more pages | - |
| DeepSeek-V4-Pro (Preview)DeepSeek, released 2026-04-24 | 72.1% | DeepSeek, DeepSeek-V4-Flash-0731 model cardchecked 2026-10-15; printed on 4 more pages | - |
| DeepSeek-V4-Flash-0731DeepSeek | 82.7% | DeepSeek, DeepSeek-V4-Flash-0731 model cardchecked 2026-10-15; printed on 3 more pages | - |
| DeepSeek-V4-Pro-0813DeepSeek | 87.9% | DeepSeek, DeepSeek-V4-Pro-0813 model cardchecked 2026-10-15; printed on 2 more pages | - |
| DeepSeek-V4-Vision-ExpDeepSeek | 83.9% | Z.ai, Z.ai GLM-5.3-Flash model cardchecked 2026-10-15 | Harness: Claude Code 2.1.207 |
| Qwen3.8-MaxAlibaba, released 2026-08-03 | 86.6% | Alibaba, Qwen3.8-Max comparisonchecked 2026-10-14; printed on 1 more page | - |
| Qwen3.7-PlusAlibaba, released 2026-06-01 | 64% | Alibaba, Qwen3.8-27B model cardchecked 2026-10-15 | Harness: Terminus |
| Qwen3.6-27BAlibaba, released 2026-04-22 | 63.4% | Alibaba, Qwen3.8-27B model cardchecked 2026-10-15 | Harness: Terminus |
| Qwen3.6-35B-A3BAlibaba, released 2026-04-16 | 44.38% | NVIDIA, NVIDIA Nemotron 3.5 Bolt 74B-A9B model cardchecked 2026-10-19; printed on 1 more page | - |
| Qwen3.5-122B-A10BAlibaba, released 2026-02-24 | 48.7% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| Qwen3.5-397B-A17BAlibaba, released 2026-02-16 | 49.9% | NVIDIA, NVIDIA Nemotron 3 Ultra 550B-A55B model cardchecked 2026-10-15 | - |
| Qwen3.7-MaxAlibaba | 74.5% | Alibaba, Qwen3.8-Max comparisonchecked 2026-10-14; printed on 1 more page | - |
| Qwen3.8-2.4T-A95BAlibaba | 82% | Meta, Meta Muse Spark Open 816B-A42B model cardchecked 2026-10-19 | Effort: xhigh; Harness: Terminus 2 |
| Qwen3.8-27BAlibaba | 73% | Alibaba, Qwen3.8-27B model cardchecked 2026-10-15 | Harness: Terminus |
| Qwen3.8-Flash-NextAlibaba | 85.8% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| Kimi K3Moonshot AI, released 2026-07-16 | 88.3% | Moonshot AI, Kimi K3 model cardchecked 2026-10-15; printed on 4 more pages | Effort: max; Harness: Kimi Code |
| Kimi K2.6Moonshot AI, released 2026-04-20 | 67.2% | NVIDIA, NVIDIA Nemotron 3 Ultra 550B-A55B model cardchecked 2026-10-15; printed on 1 more page | - |
| Kimi K2.5Moonshot AI, released 2026-01-27 | 51.3% | Thinking Machines, Thinking Machines Inkling model cardchecked 2026-10-15 | Harness: best harness per model, as printed |
| GLM-5.3-FlashZ.ai, released 2026-08-26 | 84.3% | Z.ai, Z.ai GLM-5.3-Flash model cardchecked 2026-10-15; printed on 1 more page | Effort: max; Harness: Claude Code 2.1.207 |
| GLM-5.3Z.ai, released 2026-08-18 | 88.2% | Z.ai, GLM-5.3 model cardchecked 2026-10-15; printed on 2 more pages | Effort: max; Harness: Claude Code 2.1.207 |
| GLM-5.2Z.ai, released 2026-06-16 | 82.7% | Z.ai, GLM-5.2 model cardchecked 2026-10-15; printed on 7 more pages | - |
| 81% | Z.ai, GLM-5.2 model cardchecked 2026-10-15 | Harness: Terminus-2; Context: 256000 | |
| GLM-5.1Z.ai, released 2026-04-07 | 69% | Z.ai, GLM-5.2 model cardchecked 2026-10-15; printed on 1 more page | - |
| 63.5% | Z.ai, GLM-5.2 model cardchecked 2026-10-15 | Harness: Terminus-2; Context: 256000 | |
| MiniMax M3MiniMax, released 2026-06-01 | 66% | MiniMax, MiniMax M3checked 2026-10-15; printed on 2 more pages | Harness: Terminus 2 |
| MiniMax M2.7MiniMax, released 2026-03-18 | 51.1% | MiniMax, MiniMax M3checked 2026-10-15; printed on 1 more page | Harness: Terminus 2 |
| Nemotron 3.5 Bolt 74B-A9BNVIDIA, released 2026-10-15 | 50.98% | NVIDIA, NVIDIA Nemotron 3.5 Bolt 74B-A9B model cardchecked 2026-10-19 | Model as printed: Nemotron 3.5 Bolt 74B-A9B-24GB (IQ2) |
| 53.93% | NVIDIA, NVIDIA Nemotron 3.5 Bolt 74B-A9B model cardchecked 2026-10-19 | Model as printed: Nemotron 3.5 Bolt 74B-A9B (BF16) | |
| Nemotron 3.5 Super VL 120B-A12BNVIDIA, released 2026-10-15 | 60.67% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| Nemotron 3.5 Lightning 30B-A3BNVIDIA, released 2026-08-11 | 23.46% | NVIDIA, NVIDIA Nemotron 3.5 Bolt 74B-A9B model cardchecked 2026-10-19; printed on 1 more page | - |
| Nemotron 3 UltraNVIDIA, released 2026-06-04 | 54.8% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19; printed on 4 more pages | - |
| Nemotron 3 SuperNVIDIA, released 2026-03-11 | 39.61% | NVIDIA, NVIDIA Nemotron 3.5 Lightning 30B-A3B model cardchecked 2026-10-19 | - |
| Nemotron 3 Nano 30B-A3BNVIDIA, released 2025-12-15 | 8.29% | NVIDIA, NVIDIA Nemotron 3.5 Lightning 30B-A3B model cardchecked 2026-10-19 | - |
| MiMo-V2.6-FlashXiaomi, released 2026-09-21 | 87.6% | Xiaomi, Xiaomi MiMo-V2.6-Pro-RL model cardchecked 2026-10-19 | - |
| MiMo-V2.6-ProXiaomi, released 2026-09-21 | 89.9% | Xiaomi, Xiaomi MiMo-V2.6-Pro-RL model cardchecked 2026-10-19 | - |
| MiMo-V2.5-ProXiaomi, released 2026-04-27 | 65.2% | Xiaomi, Xiaomi MiMo-V2.6-Pro-RL model cardchecked 2026-10-19 | - |
| InklingThinking Machines, released 2026-07-14 | 63.8% | Thinking Machines, Thinking Machines Inkling model cardchecked 2026-10-15; printed on 2 more pages | Effort: 0.99; Harness: best harness per model, as printed |
| Inkling-SmallThinking Machines | 53.4% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
Terminal-Bench 2.0
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| GPT-5.5OpenAI, released 2026-04-23 | 82.7% | OpenAI, Introducing GPT-5.5checked 2026-10-15 | Effort: xhigh |
| GPT-5.4 miniOpenAI, released 2026-03-17 | 60% | OpenAI, Introducing GPT-5.4 mini and nanochecked 2026-10-15 | Effort: xhigh |
| GPT-5.4 nanoOpenAI, released 2026-03-17 | 46.3% | OpenAI, Introducing GPT-5.4 mini and nanochecked 2026-10-15 | Effort: xhigh |
| GPT-5.4OpenAI, released 2026-03-05 | 75.1% | OpenAI, Introducing GPT-5.4checked 2026-10-15; printed on 7 more pages | Effort: xhigh |
| GPT-5.3-CodexOpenAI, released 2026-02-05 | 77.3% | OpenAI, Introducing GPT-5.3-Codexchecked 2026-10-15; printed on 2 more pages | Effort: xhigh |
| GPT-5.2-CodexOpenAI, released 2025-12-18 | 64% | OpenAI, Introducing GPT-5.2-Codexchecked 2026-10-15; printed on 1 more page | - |
| GPT-5.2OpenAI, released 2025-12-11 | 62.2% | OpenAI, Introducing GPT-5.2-Codexchecked 2026-10-15; printed on 8 more pages | - |
| GPT-5.1-Codex-MaxOpenAI, released 2025-11-19 | 58.1% | OpenAI, Introducing GPT-5.2-Codexchecked 2026-10-15; printed on 1 more page | - |
| GPT-5.1OpenAI, released 2025-11-12 | 47.6% | Anthropic, Claude Opus 4.5checked 2026-10-15; printed on 1 more page | - |
| 48.6% | Anthropic, Claude Opus 4.5checked 2026-10-15 | Harness: Terminus-2 in the Anthropic improved hosting environment | |
| GPT-5OpenAI, released 2025-08-07 | 35.2% | DeepSeek, DeepSeek-V3.2 model cardchecked 2026-10-15; printed on 1 more page | Effort: high |
| GPT-5 miniOpenAI, released 2025-08-07 | 38.2% | OpenAI, Introducing GPT-5.4 mini and nanochecked 2026-10-15; printed on 1 more page | Effort: high |
| gpt-oss-120bOpenAI, released 2025-08-05 | 18.7% | Alibaba, Qwen3.5-122B-A10B model cardchecked 2026-10-15 | - |
| Claude Opus 4.7Anthropic, released 2026-04-16 | 69.4% | Anthropic, Claude Opus 4.7checked 2026-10-15; printed on 1 more page | Harness: Terminus-2, thinking disabled |
| Claude Mythos PreviewAnthropic, released 2026-04-07 | 82% | Anthropic, Claude Opus 4.7checked 2026-10-15 | Harness: Terminus-2, thinking disabled |
| Claude Sonnet 4.6Anthropic, released 2026-02-17 | 59.1% | Anthropic, Claude Sonnet 4.6checked 2026-10-15; printed on 2 more pages | Effort: none; Harness: Terminus-2 (reproduced by Anthropic) |
| Claude Opus 4.6Anthropic, released 2026-02-05 | 65.4% | Anthropic, Claude Opus 4.6checked 2026-10-15; printed on 7 more pages | Harness: Terminus-2 (reproduced by Anthropic) |
| Claude Opus 4.5Anthropic, released 2025-11-24 | 59.3% | Anthropic, Claude Opus 4.5checked 2026-10-15; printed on 6 more pages | Effort: high |
| Claude Sonnet 4.5Anthropic, released 2025-09-29 | 50% | Anthropic, Claude Opus 4.5checked 2026-10-15; printed on 4 more pages | - |
| Claude Opus 4.1Anthropic, released 2025-08-05 | 46.5% | Anthropic, Claude Opus 4.5checked 2026-10-15 | - |
| Gemma 4 26B A4BGoogle, released 2026-04-02 | 34.2% | Alibaba, Qwen3.6-35B-A3B model cardchecked 2026-10-15 | Harness: Harbor/Terminus-2; Context: 256000 |
| Gemma 4 31BGoogle, released 2026-04-02 | 42.9% | Alibaba, Qwen3.6-27B model cardchecked 2026-10-15; printed on 1 more page | Harness: Harbor/Terminus-2; Context: 256000 |
| Gemini 3.1 Pro PreviewGoogle, released 2026-02-19 | 68.5% | Google, Gemini 3.1 Pro model cardchecked 2026-10-15; printed on 5 more pages | Effort: high; Harness: Terminus-2 |
| Gemini 3 ProGoogle, released 2025-11-18 | 56.9% | Google, Gemini 3.1 Pro model cardchecked 2026-10-15; printed on 8 more pages | Effort: high; Harness: Terminus-2 |
| DeepSeek-V4-Flash (Preview)DeepSeek, released 2026-04-24 | 56.6% | DeepSeek, DeepSeek-V4-Pro (preview) model cardchecked 2026-10-15 | Effort: high |
| 56.9% | DeepSeek, DeepSeek-V4-Pro (preview) model cardchecked 2026-10-15 | Effort: max | |
| 49.1% | DeepSeek, DeepSeek-V4-Pro (preview) model cardchecked 2026-10-15 | Effort: non-think | |
| DeepSeek-V4-Pro (Preview)DeepSeek, released 2026-04-24 | 63.3% | DeepSeek, DeepSeek-V4-Pro (preview) model cardchecked 2026-10-15; printed on 1 more page | Effort: high |
| 59.1% | DeepSeek, DeepSeek-V4-Pro (preview) model cardchecked 2026-10-15 | Effort: non-think | |
| 67.9% | DeepSeek, DeepSeek-V4-Pro (preview) model cardchecked 2026-10-15 | Effort: max | |
| DeepSeek-V3.2DeepSeek, released 2025-12-01 | 46.4% | DeepSeek, DeepSeek-V3.2 model cardchecked 2026-10-15; printed on 5 more pages | Effort: thinking |
| Qwen3.6-27BAlibaba, released 2026-04-22 | 59.3% | Alibaba, Qwen3.6-27B model cardchecked 2026-10-15 | Harness: Harbor/Terminus-2; Context: 256000 |
| Qwen3.6-35B-A3BAlibaba, released 2026-04-16 | 51.5% | Alibaba, Qwen3.6-35B-A3B model cardchecked 2026-10-15; printed on 1 more page | Harness: Harbor/Terminus-2; Context: 256000 |
| Qwen3.6-PlusAlibaba, released 2026-04-02 | 61.6% | Z.ai, GLM-5.1 model cardchecked 2026-10-15 | Harness: Terminus-2 |
| Qwen3.5-122B-A10BAlibaba, released 2026-02-24 | 49.4% | Alibaba, Qwen3.5-122B-A10B model cardchecked 2026-10-15 | - |
| Qwen3.5-27BAlibaba, released 2026-02-24 | 41.6% | Alibaba, Qwen3.5-122B-A10B model cardchecked 2026-10-15; printed on 2 more pages | - |
| Qwen3.5-35B-A3BAlibaba, released 2026-02-24 | 40.5% | Alibaba, Qwen3.5-122B-A10B model cardchecked 2026-10-15; printed on 1 more page | - |
| Qwen3.5-397B-A17BAlibaba, released 2026-02-16 | 52.5% | Alibaba, Qwen3.5-397B-A17B model cardchecked 2026-10-15; printed on 1 more page | - |
| Qwen3-Max-ThinkingAlibaba | 22.5% | Alibaba, Qwen3.5-397B-A17B model cardchecked 2026-10-15 | - |
| Kimi K2.6Moonshot AI, released 2026-04-20 | 66.7% | Moonshot AI, Kimi K2.6 model cardchecked 2026-10-15; printed on 2 more pages | Harness: Terminus-2 |
| Kimi K2.5Moonshot AI, released 2026-01-27 | 50.8% | Moonshot AI, Kimi K2.5 model cardchecked 2026-10-15; printed on 4 more pages | Harness: Terminus-2 |
| Kimi K2 ThinkingMoonshot AI, released 2025-11-06 | 35.7% | Z.ai, GLM-4.7 model cardchecked 2026-10-15 | - |
| GLM-5.1Z.ai, released 2026-04-07 | 63.5% | Z.ai, GLM-5.1 model cardchecked 2026-10-15; printed on 2 more pages | Harness: Terminus-2 |
| 69% | Z.ai, GLM-5.1 model cardchecked 2026-10-15 | Harness: Claude Code | |
| GLM-5Z.ai, released 2026-02-12 | 56.2% | Z.ai, GLM-5 model cardchecked 2026-10-15; printed on 1 more page | Harness: Claude Code 2.1.14 |
| 56.2% | Z.ai, GLM-5 model cardchecked 2026-10-15 | Harness: Terminus 2 | |
| GLM-4.7Z.ai, released 2025-12-22 | 41% | Z.ai, GLM-4.7 model cardchecked 2026-10-15; printed on 1 more page | - |
| GLM-4.6Z.ai, released 2025-09-30 | 24.5% | Z.ai, GLM-4.7 model cardchecked 2026-10-15 | - |
| MiniMax M2.7MiniMax, released 2026-03-18 | 57% | MiniMax, MiniMax-M2.7 model cardchecked 2026-10-15; printed on 1 more page | - |
| MAI-Thinking-1Microsoft, released 2026-08-12 | 46% | Microsoft, Microsoft AI: Introducing MAI-Thinking-1checked 2026-10-19 | - |
Terminal-Bench 4.0
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| GPT-6.1 TerraOpenAI, released 2026-10-26 | 50.3% | OpenAI, GPT-6.1 Terrachecked 2026-10-26 | - |
| GPT-6 LunaOpenAI, released 2026-09-22 | 16.4% | Anthropic, Claude Haiku 5.5checked 2026-10-15 | - |
| GPT-6 AstraOpenAI, released 2026-09-03 | 57.9% | OpenAI, GPT-6 Astrachecked 2026-10-14; printed on 7 more pages | - |
| GPT-5.6 SolOpenAI, released 2026-07-09 | 37.3% | OpenAI, GPT-6 Astrachecked 2026-10-14; printed on 10 more pages | - |
| GPT-5.6 TerraOpenAI, released 2026-07-09 | 23.6% | Google, Gemini 3.8 Flash model cardchecked 2026-10-15; printed on 1 more page | - |
| Claude Haiku 5.5Anthropic, released 2026-10-07 | 39.2% | Anthropic, Claude Haiku 5.5checked 2026-10-15; printed on 1 more page | - |
| Claude Sonnet 5.5Anthropic, released 2026-09-28 | 70.6% | Anthropic, Claude Sonnet 5.5checked 2026-10-14; printed on 3 more pages | - |
| Claude Opus 5.5Anthropic, released 2026-09-22 | 66.4% | Anthropic, Claude Opus 5.5checked 2026-10-14; printed on 8 more pages | - |
| Claude Fable 5.1Anthropic, released 2026-09-01 | 55.8% | Anthropic, Claude Fable 5.1 and Mythos 5.1checked 2026-10-15; printed on 11 more pages | Model setup as reported: Release table; exact effort is unspecified here unless stated. See system card for full protocol. |
| 60.9% | Anthropic, Claude Fable 5.1 and Mythos 5.1checked 2026-10-15 | Model setup as reported: Second value printed in the Fable 5.1 cell as "60.9% (Mythos 5.1)": filed under the Fable 5.1 heading, the label names Claude Mythos 5.1. | |
| Claude Mythos 5.1Anthropic, released 2026-09-01 | 60.9% | Anthropic, System Card: Claude Fable 5.1 & Claude Mythos 5.1checked 2026-10-25; printed on 2 more pages | - |
| Claude Opus 5Anthropic, released 2026-07-24 | 52.3% | Anthropic, Claude Opus 5.5checked 2026-10-14; printed on 10 more pages | - |
| Claude Sonnet 5Anthropic, released 2026-06-30 | 10.3% | Anthropic, Claude Sonnet 5.5checked 2026-10-14; printed on 2 more pages | - |
| Claude Fable 5Anthropic, released 2026-06-09 | 42% | Anthropic, Claude Fable 5.1 and Mythos 5.1checked 2026-10-15; printed on 3 more pages | - |
| Claude Haiku 4.5Anthropic, released 2025-10-15 | 0% | Anthropic, Claude Haiku 5.5checked 2026-10-15; printed on 1 more page | - |
| Gemini 4 CarbonGoogle, released 2026-10-28 | 65.2% | Google, Gemini 4 Carbonchecked 2026-10-29 | - |
| Gemini 4 ArgonGoogle, released 2026-09-30 | 57.4% | Google, Gemini 4 Argonchecked 2026-10-29; printed on 1 more page | - |
| Gemini 3.8 FlashGoogle, released 2026-09-02 | 19.1% | Google, Gemini 3.8 Flash model cardchecked 2026-10-15; printed on 1 more page | - |
| Gemini 3.7 FlashGoogle, released 2026-08-13 | 11.2% | Google, Gemini 3.8 Flash model cardchecked 2026-10-15 | - |
| Grok 4.7xAI, released 2026-09-21 | 37.6% | xAI, Grok 4.7checked 2026-10-25; printed on 1 more page | Effort: xhigh |
| Grok 4.6xAI, released 2026-08-12 | 20.3% | xAI, Grok 4.7checked 2026-10-25; printed on 1 more page | Effort: high |
| Grok 4.5xAI, released 2026-07-16 | 12.4% | xAI, Grok 4.7 model cardchecked 2026-10-14 | Effort: high |
| Muse WonderMeta, released 2026-10-12 | 52% | Meta, Meta Muse Wonder model cardchecked 2026-10-19 | Effort: max |
| Muse Wonder FullMeta, released 2026-10-12 | 57.1% | Meta, Meta Muse Wonder model cardchecked 2026-10-19 | Effort: max |
| DeepSeek-V4.1-Flash-1017DeepSeek, released 2026-10-17 | 42.3% | DeepSeek, DeepSeek-V4.1-Flash-1017 model cardchecked 2026-10-18; printed on 2 more pages | Harness: Claude Code |
| DeepSeek-V4.1-FlashDeepSeek, released 2026-09-10 | 31.2% | DeepSeek, DeepSeek-V4.1-Flash model cardchecked 2026-10-15; printed on 2 more pages | Effort: max; Harness: DeepSeek Harness (DSH) Minimal; Context: 1M |
| DeepSeek-V4-Flash-0731DeepSeek | 7% | DeepSeek, DeepSeek-V4.1-Flash model cardchecked 2026-10-15; printed on 1 more page | Effort: max |
| DeepSeek-V4-Pro-0813DeepSeek | 12.4% | DeepSeek, DeepSeek-V4.1-Flash model cardchecked 2026-10-15; printed on 1 more page | Effort: max |
| Qwen3.8-MaxAlibaba, released 2026-08-03 | 17% | Mistral AI, Mistral Large 4 (preview)checked 2026-10-15 | - |
| Qwen3.5-122B-A10BAlibaba, released 2026-02-24 | 0% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| Qwen3.8-Flash-NextAlibaba | 20.71% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| Kimi K3.1 ProMoonshot AI, released 2026-10-23 | 65.7% | Moonshot AI, Introducing Kimi K3.1 Prochecked 2026-10-24 | Effort: max; Harness: Claude Code |
| Kimi K3Moonshot AI, released 2026-07-16 | 12.6% | DeepSeek, DeepSeek-V4.1-Flash model cardchecked 2026-10-15; printed on 3 more pages | - |
| GLM-5.3-FlashZ.ai, released 2026-08-26 | 15.15% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| GLM-5.3Z.ai, released 2026-08-18 | 37.9% | DeepSeek, DeepSeek-V4.1-Flash model cardchecked 2026-10-15; printed on 4 more pages | - |
| Mistral Large 4 (preview)Mistral AI, released 2026-10-06 | 28.3% | Mistral AI, Mistral Large 4 (preview)checked 2026-10-15 | - |
| MiniMax-M3.1-FlashMiniMax, released 2026-10-19 | 38.1% | MiniMax, MiniMax-M3.1-Flash model cardchecked 2026-10-25 | Harness: Codex 0.149.1 |
| Nemotron 3.5 Super VL 120B-A12BNVIDIA, released 2026-10-15 | 1% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| Nemotron 3 UltraNVIDIA, released 2026-06-04 | 0% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| MiMo-V2.6-FlashXiaomi, released 2026-09-21 | 28.8% | Xiaomi, Xiaomi MiMo-V2.6-Pro-RL model cardchecked 2026-10-19 | - |
| MiMo-V2.6-ProXiaomi, released 2026-09-21 | 34.9% | Xiaomi, Xiaomi MiMo-V2.6-Pro-RL model cardchecked 2026-10-19 | - |
| MiMo-V2.5-ProXiaomi, released 2026-04-27 | 1.5% | Xiaomi, Xiaomi MiMo-V2.6-Pro-RL model cardchecked 2026-10-19 | - |
| Inkling-SmallThinking Machines | 0.5% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
Terminal-Bench (version not stated)
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| GPT-5OpenAI, released 2025-08-07 | 43.8% | Anthropic, Claude Haiku 4.5checked 2026-10-15; printed on 2 more pages | - |
| o3OpenAI, released 2025-04-16 | 30.2% | Anthropic, Claude Opus 4 and Claude Sonnet 4checked 2026-10-15; printed on 1 more page | - |
| GPT-4.1OpenAI, released 2025-04-14 | 30.3% | Anthropic, Claude Opus 4 and Claude Sonnet 4checked 2026-10-15; printed on 1 more page | - |
| Claude Haiku 4.5Anthropic, released 2025-10-15 | 41% | Anthropic, Claude Haiku 4.5checked 2026-10-15 | Effort: thinking budget 128K |
| Claude Sonnet 4.5Anthropic, released 2025-09-29 | 50% | Anthropic, Claude Sonnet 4.5checked 2026-10-15; printed on 2 more pages | - |
| Claude Opus 4.1Anthropic, released 2025-08-05 | 43.3% | Anthropic, Claude Opus 4.1checked 2026-10-15; printed on 1 more page | Effort: none; Harness: Terminus 1 |
| Claude Opus 4Anthropic, released 2025-05-22 | 50% | Anthropic, Claude Opus 4 and Claude Sonnet 4checked 2026-10-15; printed on 2 more pages | Effort: none; Attempt budget: parallel test-time compute: multiple samples, the best selected by an internal scoring model |
| 43.2% | Anthropic, Claude Opus 4 and Claude Sonnet 4checked 2026-10-15 | Effort: none | |
| 39.2% | Anthropic, Claude Opus 4 and Claude Sonnet 4checked 2026-10-15 | Effort: none; Harness: the same agent as the non-Claude models | |
| Claude Sonnet 4Anthropic, released 2025-05-22 | 36.4% | Anthropic, Claude Sonnet 4.5checked 2026-10-15; printed on 4 more pages | - |
| Claude 3.7 SonnetAnthropic, released 2025-02-24 | 35.2% | Anthropic, Claude Opus 4 and Claude Sonnet 4checked 2026-10-15 | - |
| Gemini 2.5 FlashGoogle, released 2025-05-20 | 16.8% | Moonshot AI, Kimi K2 release pagechecked 2026-10-15 | Harness: Terminus |
| Gemini 2.5 ProGoogle, released 2025-03-25 | 25.3% | Anthropic, Claude Opus 4 and Claude Sonnet 4checked 2026-10-15; printed on 3 more pages | - |
| DeepSeek-V3.2-ExpDeepSeek, released 2025-09-29 | 37.7% | DeepSeek, DeepSeek-V3.2-Exp model cardchecked 2026-10-15; printed on 1 more page | Tools: enabled; exact tools in source |
| DeepSeek-V3.1-TerminusDeepSeek, released 2025-09-22 | 36.7% | DeepSeek, DeepSeek-V3.1-Terminus model cardchecked 2026-10-15; printed on 1 more page | Tools: enabled; exact tools in source |
| DeepSeek-V3.1DeepSeek, released 2025-08-21 | 31.3% | DeepSeek, DeepSeek-V3.1-Terminus model cardchecked 2026-10-15 | Tools: enabled; exact tools in source |
| DeepSeek-V3-0324DeepSeek, released 2025-03-24 | 16.3% | Moonshot AI, Kimi K2 release pagechecked 2026-10-15 | Harness: Terminus |
| Qwen3-235B-A22BAlibaba, released 2025-04-29 | 6.6% | Moonshot AI, Kimi K2 release pagechecked 2026-10-15 | Effort: non-thinking; Harness: Terminus |
| Kimi K2 ThinkingMoonshot AI, released 2025-11-06 | 47.1% | Moonshot AI, Kimi K2 Thinking release pagechecked 2026-10-15 | Effort: thinking; Harness: Terminus-2 with the provided JSON parser; Tools: enabled; simulated tools (JSON) |
| Kimi K2 Instruct 0905Moonshot AI, released 2025-09-05 | 44.5% | Moonshot AI, Kimi K2 Thinking release pagechecked 2026-10-15 | Harness: Terminus-2 with the provided JSON parser; Tools: enabled; simulated tools (JSON) |
| Kimi K2 InstructMoonshot AI, released 2025-07-11 | 30% | Moonshot AI, Kimi K2 release pagechecked 2026-10-15 | Harness: in-house framework |
| 25% | Moonshot AI, Kimi K2 release pagechecked 2026-10-15 | Harness: Terminus |
Terminal-Bench 3.0
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| GPT-5.6 SolOpenAI, released 2026-07-09 | 34.4% | DeepSeek, DeepSeek-V4.1-Flash model cardchecked 2026-10-15; printed on 2 more pages | - |
| GPT-5.6 TerraOpenAI, released 2026-07-09 | 20.8% | Google, Gemini 3.7 Flash model cardchecked 2026-10-15 | - |
| Claude Opus 5Anthropic, released 2026-07-24 | 43.3% | DeepSeek, DeepSeek-V4.1-Flash model cardchecked 2026-10-15 | - |
| Claude Sonnet 5Anthropic, released 2026-06-30 | 14.6% | Google, Gemini 3.7 Flash model cardchecked 2026-10-15 | - |
| Claude Fable 5Anthropic, released 2026-06-09 | 34.1% | xAI, Introducing Grok 4.6checked 2026-10-15; printed on 1 more page | Effort: max |
| Claude Opus 4.8Anthropic, released 2026-05-28 | 21.1% | Z.ai, GLM-5.3 model cardchecked 2026-10-15 | - |
| Gemini 3.7 FlashGoogle, released 2026-08-13 | 14.9% | Google, Gemini 3.7 Flash model cardchecked 2026-10-15 | - |
| Gemini 3.6 FlashGoogle, released 2026-07-21 | 5.4% | Google, Gemini 3.7 Flash model cardchecked 2026-10-15 | - |
| Grok 4.6xAI, released 2026-08-12 | 26% | xAI, Introducing Grok 4.6checked 2026-10-15 | Effort: high |
| Grok 4.5xAI, released 2026-07-16 | 15.7% | xAI, Introducing Grok 4.6checked 2026-10-15 | Effort: high |
| DeepSeek-V4.1-FlashDeepSeek, released 2026-09-10 | 30% | DeepSeek, DeepSeek-V4.1-Flash model cardchecked 2026-10-15 | Effort: max; Harness: DeepSeek Harness (DSH) Minimal; Context: 1M |
| DeepSeek-V4-Flash-0731DeepSeek | 7.6% | DeepSeek, DeepSeek-V4.1-Flash model cardchecked 2026-10-15 | Effort: max |
| DeepSeek-V4-Pro-0813DeepSeek | 11.8% | DeepSeek, DeepSeek-V4.1-Flash model cardchecked 2026-10-15 | Effort: max |
| Kimi K3Moonshot AI, released 2026-07-16 | 17.7% | DeepSeek, DeepSeek-V4.1-Flash model cardchecked 2026-10-15; printed on 1 more page | - |
| GLM-5.3Z.ai, released 2026-08-18 | 28.3% | Z.ai, GLM-5.3 model cardchecked 2026-10-15; printed on 1 more page | Effort: max; Harness: Claude Code 2.1.207; Context: 400000 |
| GLM-5.2Z.ai, released 2026-06-16 | 4.6% | Z.ai, GLM-5.3 model cardchecked 2026-10-15 | - |
Terminal-Bench Hard
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| GPT-5.1OpenAI, released 2025-11-12 | 43% | Z.ai, GLM-4.7 model cardchecked 2026-10-15 | Effort: high |
| GPT-5OpenAI, released 2025-08-07 | 30.5% | Z.ai, GLM-4.7 model cardchecked 2026-10-15 | Effort: high |
| gpt-oss-20bOpenAI, released 2025-08-05 | 6% | NVIDIA, NVIDIA Nemotron 3 Nano 30B-A3B model cardchecked 2026-10-15 | - |
| Claude Sonnet 4.5Anthropic, released 2025-09-29 | 33.3% | Z.ai, GLM-4.7 model cardchecked 2026-10-15 | - |
| Gemini 3 ProGoogle, released 2025-11-18 | 39% | Z.ai, GLM-4.7 model cardchecked 2026-10-15 | - |
| DeepSeek-V3.2DeepSeek, released 2025-12-01 | 35.4% | Z.ai, GLM-4.7 model cardchecked 2026-10-15 | - |
| Qwen3-30B-A3B-Thinking-2507Alibaba, released 2025-07-30 | 5% | NVIDIA, NVIDIA Nemotron 3 Nano 30B-A3B model cardchecked 2026-10-15 | Effort: thinking |
| Kimi K2 ThinkingMoonshot AI, released 2025-11-06 | 30.6% | Z.ai, GLM-4.7 model cardchecked 2026-10-15 | - |
| GLM-4.7Z.ai, released 2025-12-22 | 33.3% | Z.ai, GLM-4.7 model cardchecked 2026-10-15 | - |
| GLM-4.6Z.ai, released 2025-09-30 | 23.6% | Z.ai, GLM-4.7 model cardchecked 2026-10-15 | - |
| Nemotron 3 Nano 30B-A3BNVIDIA, released 2025-12-15 | 8.5% | NVIDIA, NVIDIA Nemotron 3 Nano 30B-A3B model cardchecked 2026-10-15 | - |
Terminal-Bench 2.0 (Z.ai verified set)
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| GLM-5Z.ai, released 2026-02-12 | 61.1% | Z.ai, GLM-5 model cardchecked 2026-10-15 | Harness: Claude Code 2.1.14 |
| 60.7% | Z.ai, GLM-5 model cardchecked 2026-10-15 | Harness: Terminus 2 |
Every number here is read from the same catalog file as the benchmark table, data version 2026103001, and is the number the table shows for that model and test. Where one page printed a model under several setups, each value is listed. The catalog is public in armanobosyan/sugra-llm-benchmarks; a wrong number can be reported from the table, with Report an error under the value, or as an issue in that repository.