Benchmarks · coding
Terminal-Bench v2.1-AA
Terminal-Bench v2.1 run by AA: agentic command-line software engineering tasks; task pass rate.
85 models · 142 results · data as of 2026-10-29 · artificialanalysis.ai ↗
Leaderboard
Fable 5.1 max
More from this benchmark
4 published measuresScore vs any published measure4 measures: sub-scores, splits, counts
Breakdown
darker is better within each column · click a column to sort| Model | ||||
|---|---|---|---|---|
| Fable 5.1 max, fallback | 149,743,249 | 1,331,788 | 4,019,111 | 142,508,037 |
| GPT-6 Astra high | 13,715,507 | 362,929 | 152,128 | 12,237,045 |
| GPT-5.6 Sol xhigh | 18,963,277 | 516,757 | 442,910 | 17,507,646 |
| Opus 5 max | 132,353,877 | 1,333,842 | 3,102,688 | 126,095,036 |
| Grok 4.6 high | 75,473,643 | 697,912 | 992,376 | 73,850,969 |
| GPT-5.6 Terra max | 74,758,292 | 799,505 | 1,522,272 | 71,665,427 |
| Gemini 3.8 Flash high | 283,570,242 | 1,697,849 | 2,956,103 | 276,083,298 |
| Gemini 3.7 Flash high | 190,403,003 | 1,748,588 | 1,536,517 | 184,600,851 |
| Muse Spark 1.3 xhigh | 286,758,029 | 1,194,905 | 2,885,142 | 279,716,694 |
| Kimi K3 max | 52,251,794 | 525,087 | 1,420,531 | 49,461,474 |
| Opus 4.8 max | 161,373,863 | 1,263,528 | 5,626,763 | 152,885,621 |
| GLM 5.3 Flash max | 157,758,836 | 889,979 | 2,731,018 | 151,446,258 |
| GPT-5.5 xhigh | 26,929,862 | 617,745 | 2,100,153 | 25,720,125 |
| GLM 5.3 max | 136,692,446 | 853,063 | 2,698,125 | 130,793,242 |
| Opus 4.7 max | 245,109,205 | 1,241,933 | 4,528,964 | 237,147,932 |
| Qwen 3.8 2.4T-A95B | 109,664,633 | 889,247 | 2,441,444 | 104,828,681 |
| Grok 4.5 high | 57,949,273 | 897,744 | 546,025 | 56,403,230 |
| GPT-5.6 Luna max | 82,569,478 | 887,304 | 1,259,382 | 81,004,211 |
| Sonnet 5 max | 297,414,309 | 1,995,615 | 5,851,108 | 287,056,095 |
| Muse Spark 1.2 xhigh | 245,820,870 | 1,326,802 | 2,784,057 | 242,865,838 |
| Qwen 3.8 27B xhigh | 244,264,574 | 1,357,477 | 4,654,402 | 233,189,640 |
| Gemini 3.5 Flash high | 149,685,552 | 1,012,505 | 1,839,754 | 143,690,018 |
| GPT-5.4 xhigh | 21,780,476 | 786,324 | 6,527,094 | 20,289,065 |
| GLM 5.2 max | 78,240,101 | 1,097,099 | 6,166,636 | 73,682,807 |
| Muse Spark 1.1 xhigh | 209,795,318 | 1,383,505 | 2,043,020 | 203,789,293 |
All results
142 rows| # | Model | Effort | Terminal-Bench v2.1-AA |
|---|---|---|---|
| 1 | Fable 5.1 | max | 91.4% |
| 2 | Fable 5.1 | xhigh | 91.0% |
| 3 | Fable 5.1 | high | 89.9% |
| 4 | GPT-6 Astra | high | 89.9% |
| 5 | GPT-5.6 Sol | xhigh | 89.5% |
| 6 | GPT-6 Astra | medium | 89.5% |
| 7 | Opus 5 | max | 89.1% |
| 8 | GPT-6 Astra | xhigh | 89.1% |
| 9 | GPT-6 Astra | max | 88.4% |
| 10 | Grok 4.6 | high | 88.4% |
| 11 | Fable 5.1 | medium | 88.0% |
| 12 | Opus 5 | xhigh | 88.0% |
| 13 | GPT-5.6 Sol | max | 88.0% |
| 14 | GPT-5.6 Terra | max | 88.0% |
| 15 | GPT-6 Astra | low | 88.0% |
| 16 | Grok 4.6 | xhigh | 88.0% |
| 17 | Opus 5 | high | 87.6% |
| 18 | Gemini 3.8 Flash | high | 87.6% |
| 19 | GPT-5.6 Sol | high | 87.3% |
| 20 | Opus 5 | medium | 86.1% |
| 21 | GPT-5.6 Sol | medium | 86.1% |
| 22 | Gemini 3.7 Flash | high | 85.8% |
| 23 | Muse Spark 1.3 | xhigh | 85.4% |
| 24 | Fable 5.1 | low | 85.0% |
| 25 | Kimi K3 | max | 85.0% |
| 26 | Opus 4.8 | max | 84.6% |
| 27 | GLM 5.3 Flash | max | 84.3% |
| 28 | GPT-5.5 | xhigh | 84.3% |
| 29 | Grok 4.6 | medium | 84.3% |
| 30 | Muse Spark 1.3 | max | 84.3% |