BenchLeader
Menu
Best LLMsChoose a modelCoding plansLeaderboardsModelsBenchmarksProvidersReleases12

Terminal-Bench 2.1 (AA) leaderboard

As of 9 Nov 2026, the top score on Terminal-Bench 2.1 (AA) is 91.4% by Claude Fable 5.1 (Anthropic), ahead of GPT-6 Astra at 89.9%, across 240 model configurations with a published result (Artificial Analysis, result dated 1 Sep 2026). The previous Terminal-Bench release, kept because far more models have been run on it than on 4.0. Terminal-Bench 2.1 (AA) is shown for reference only and does not count toward the BenchLeader Index.

Key facts (as of 9 Nov 2026)

  • Claude Fable 5.1 holds the top Terminal-Bench 2.1 (AA) score at 91.4% (Artificial Analysis, 1 Sep 2026).
  • GPT-6 Astra is second on Terminal-Bench 2.1 (AA) at 89.9% (Artificial Analysis, 3 Sep 2026).
  • Terminal-Bench 2.1 (AA): The previous Terminal-Bench release, kept because far more models have been run on it than on 4.0 (Artificial Analysis).
  • 240 model configurations have a published Terminal-Bench 2.1 (AA) result on BenchLeader (BenchLeader, 9 Nov 2026).
  • Terminal-Bench 2.1 (AA) does not feed the BenchLeader Index; it is shown for reference (BenchLeader methodology).

The previous Terminal-Bench release, kept because far more models have been run on it than on 4.0.

Published by
Artificial Analysis
Category
Coding
Index weight
Reference only
Models
240
Data as of
9 Nov 2026

Source: Artificial Analysis (artificialanalysis.ai). Data taken from the public leaderboard.

Top 15
  1. 1Claude Fable 5.1 (max)91.4%
  2. 2Claude Fable 5.1 (xhigh)91.0%
  3. 3GPT-6 Astra (high)89.9%
  4. 4Claude Fable 5.1 (high)89.9%
  5. 5GPT-6 Astra (medium)89.5%
  6. 6GPT-5.6 Sol (xhigh)89.5%
  7. 7GPT-6 Astra (xhigh)89.1%
  8. 8Claude Opus 5 (max)89.1%
  9. 9Qwen3.8 Max (max)88.8%
  10. 10GPT-6 Astra (max)88.4%
  11. 11Grok 4.6 (high)88.4%
  12. 12GPT-5.6 Sol (max)88.0%
  13. 13GPT-6 Astra (low)88.0%
  14. 14Claude Fable 5.1 (medium)88.0%
  15. 15GPT-5.6 Terra (max)88.0%
240 of 240
#
1Claude Fable 5.1max91.4%66.92026-09-01
2Claude Fable 5.1xhigh91.0%64.62026-09-01
3GPT-6 Astrahigh89.9%66.52026-09-03
4Claude Fable 5.1high89.9%63.32026-09-01
5GPT-6 Astramedium89.5%64.52026-09-03
6GPT-5.6 Solxhigh89.5%64.22026-07-09
7GPT-6 Astraxhigh89.1%66.22026-09-03
8Claude Opus 5max89.1%65.12026-07-24
9Qwen3.8 Maxmax88.8%58.92026-09-02
10GPT-6 Astramax88.4%67.42026-09-03
11Grok 4.6high88.4%60.52026-08-12
12GPT-5.6 Solmax88.0%65.42026-07-09
13GPT-6 Astralow88.0%62.22026-09-03
14Claude Fable 5.1medium88.0%61.72026-09-01
15GPT-5.6 Terramax88.0%60.52026-07-09
16Claude Opus 5xhigh88.0%60.12026-07-24
17Grok 4.6xhigh88.0%57.62026-08-12
18Claude Opus 5high87.6%64.22026-07-24
19Gemini 3.8 Flashhigh87.6%61.82026-09-02
20GPT-5.6 Solhigh87.3%63.52026-07-09
21GPT-5.6 Solmedium86.1%61.42026-07-09
22Claude Opus 5medium86.1%57.42026-07-24
23Qwen3.8-Flash-Next86.1%54.42026-08-26
24Gemini 3.7 Flashhigh85.8%62.22026-08-13
25Muse Spark 1.3xhigh85.4%58.12026-09-02
26Kimi K3max85.0%62.02026-07-16
27Claude Fable 5.1low85.0%58.72026-09-01
28Claude Fable 5max84.6%65.32026-06-09
29Claude Opus 4.8max84.6%60.22026-05-28
30GPT-5.5xhigh84.3%64.32026-04-23
31Muse Spark 1.3max84.3%61.72026-09-02
32Grok 4.6medium84.3%59.52026-08-12
33GLM 5.3 Flash84.3%56.82026-08-26
34Gemini 3.8 Flashmedium83.9%60.32026-09-02
35GLM 5.3max83.9%58.92026-08-18
36Claude Opus 4.7max83.2%59.52026-04-16
37Gemini 3.8 Flashlow83.2%55.92026-09-02
38Kimi K3low82.4%53.72026-07-16
39Qwen3.8 2.4T A95B82.0%51.62026-08-12
40Grok 4.5high81.7%56.82026-07-08
41NewInkling 2 Small81.3%58.32026-11-05
42GPT-5.6 Lunamax80.9%56.22026-07-09
43GPT-5.5medium80.5%60.92026-04-23
44Claude Sonnet 5max80.5%53.62026-06-30
45Muse Spark 1.2xhigh80.2%61.72026-08-05
46GPT-5.6 Terraxhigh80.2%60.32026-07-09
47Gemini 3.7 Flashlow79.8%57.42026-08-13
48Qwen3.8 27Bxhigh79.8%52.42026-08-14
49GPT-5.5high79.4%63.72026-04-23
50Gemini 3.5 Flashhigh78.7%62.52026-05-19
51DeepSeek V4 Flashmax78.7%57.02026-07-31
52DeepSeek V4 Promax78.7%56.92026-08-13
53GPT-5.4xhigh78.3%63.22026-03-05
54Gemini 3.7 Flashmedium78.3%60.22026-08-13
55Muse Spark 1.1xhigh77.9%59.62026-07-09
56GLM 5.2max77.9%59.42026-06-16
57GPT-5.6 Lunaxhigh77.9%58.42026-07-09
58Gemini 3.6 Flashhigh77.5%61.22026-07-21
59GPT-5.6 Sollow76.8%60.32026-07-09
60Claude Opus 5low76.4%57.02026-07-24

More coding benchmarks · All LLM benchmarks

Related benchmarks

Top head-to-heads on Terminal-Bench 2.1 (AA)

Cite as: BenchLeader, “Terminal-Bench 2.1 (AA) leaderboard”, data as of 9 Nov 2026, https://www.benchleader.com/benchmarks/aa_terminalbench_21.

Terminal-Bench 2.1 (AA): questions

What does Terminal-Bench 2.1 (AA) measure?
The previous Terminal-Bench release, kept because far more models have been run on it than on 4.0. Scores are reported in percent of tasks solved; higher is better.
What is the top score on Terminal-Bench 2.1 (AA)?
As of 9 Nov 2026, the highest score on Terminal-Bench 2.1 (AA) is 91.4% by Claude Fable 5.1 (result dated 1 Sep 2026), ahead of GPT-6 Astra at 89.9%. Scores come from Artificial Analysis and are re-read daily.
Which AI model leads Terminal-Bench 2.1 (AA)?
Claude Fable 5.1 leads Terminal-Bench 2.1 (AA) with 91.4% as of 9 Nov 2026 (result dated 1 Sep 2026), ahead of GPT-6 Astra at 89.9%.
How many models have Terminal-Bench 2.1 (AA) results?
240 model configurations have a Terminal-Bench 2.1 (AA) result on BenchLeader, all taken from Artificial Analysis.
Who runs Terminal-Bench 2.1 (AA) and how often is it updated?
Terminal-Bench 2.1 (AA) is published by Artificial Analysis. BenchLeader re-reads the published results once a day and records the date each result was published.
Does Terminal-Bench 2.1 (AA) count toward the BenchLeader Index?
No. Terminal-Bench 2.1 (AA) is shown for reference but left out of the composite index.