BenchLeader

AA-Omniscience v1.1: accuracy leaderboard

As of 15 Nov 2026, the top score on AA-Omniscience v1.1: accuracy is 79.1% by Claude Fable 5.2 (Anthropic), ahead of Claude Fable 5.1 at 68.4%, across 142 model configurations with a published result (Artificial Analysis, result dated 9 Nov 2026). The share of AA-Omniscience questions answered correctly, before any penalty for confident wrong answers. AA-Omniscience v1.1: accuracy is shown for reference only and does not count toward the BenchLeader Index.

Key facts (as of 15 Nov 2026)

  • Claude Fable 5.2 holds the top AA-Omniscience v1.1: accuracy score at 79.1% (Artificial Analysis, 9 Nov 2026).
  • Claude Fable 5.1 is second on AA-Omniscience v1.1: accuracy at 68.4% (Artificial Analysis, 1 Sep 2026).
  • AA-Omniscience v1.1: accuracy: The share of AA-Omniscience questions answered correctly, before any penalty for confident wrong answers (Artificial Analysis).
  • 142 model configurations have a published AA-Omniscience v1.1: accuracy result on BenchLeader (BenchLeader, 15 Nov 2026).
  • AA-Omniscience v1.1: accuracy does not feed the BenchLeader Index; it is shown for reference (BenchLeader methodology).

The share of AA-Omniscience questions answered correctly, before any penalty for confident wrong answers.

Published by
Artificial Analysis
Category
Knowledge
Index weight
Reference only
Models
142
Data as of
15 Nov 2026

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

142 of 142
#
1NewClaude Fable 5.2max79.1%72.72026-11-09
2NewClaude Fable 5.2xhigh78.0%69.72026-11-09
3NewClaude Fable 5.2high77.2%68.42026-11-09
4NewClaude Fable 5.2medium77.0%66.32026-11-09
5NewClaude Fable 5.2low76.3%65.32026-11-09
6Claude Fable 5.1max68.4%66.72026-09-01
7Claude Fable 5.1xhigh67.2%64.62026-09-01
8Claude Opus 5.5xhigh65.5%65.42026-09-22
9Claude Opus 5.5max65.5%68.42026-09-22
10Claude Opus 5.5high64.8%66.02026-09-22
11Claude Fable 5.1high64.7%63.72026-09-01
12Claude Opus 5.5medium63.9%61.22026-09-22
13Claude Opus 5.5low63.3%58.92026-09-22
14Claude Fable 5.1medium62.6%61.72026-09-01
15GPT-6.1 Solmax61.8%66.62026-09-29
16GPT-6 Astramax61.6%67.32026-09-03
17GPT-6 Astrahigh61.1%66.32026-09-03
18GPT-6.1 Solxhigh61.1%63.62026-09-29
19GPT-6 Astraxhigh61.0%66.02026-09-03
20Claude Opus 5max60.9%65.02026-07-24
21GPT-6.1 Solhigh60.4%63.42026-09-29
22GPT-6 Astramedium59.6%64.42026-09-03
23GPT-6.1 Solmedium59.6%62.72026-09-29
24Claude Fable 5.1low58.7%58.82026-09-01
25GPT-6.1 Sollow58.6%60.02026-09-29
26GPT-6 Astralow58.6%62.02026-09-03
27Gemini 4 Carbonhigh58.3%68.92026-10-28
28Claude Sonnet 5.5max51.5%64.22026-09-28
29GPT-6 Solmax51.1%62.62026-09-22
30Claude Sonnet 5.5xhigh50.9%62.52026-09-28
31Gemini 3.8 Flashhigh50.5%59.42026-09-02
32NewGemini 4 Hydrogenxhigh50.5%61.82026-11-12
33Claude Sonnet 5.5high50.3%59.52026-09-28
34Hy450.0%56.72026-11-03
35Gemini 3.8 Flashmedium49.6%60.22026-09-02
36Gemini 4 Carbonlow49.2%60.12026-10-28
37Gemini 4 Argonmedium49.0%61.62026-09-30
38Gemini 4 Carbonmedium48.9%65.82026-10-28
39Kimi K3.1 Pro48.9%61.82026-10-23
40NewGemini 4 Hydrogenmedium48.3%–2026-11-12
41Gemini 4 Argonhigh48.3%65.32026-09-30
42Gemini 3.8 Flashlow48.3%55.92026-09-02
43NewGemini 4 Hydrogenhigh48.2%63.42026-11-12
44GPT-6.1 Terraxhigh48.0%60.02026-10-26
45GPT-6.1 Terramax47.6%64.12026-10-26
46GPT-6.1 Terramedium47.4%58.52026-10-26
47GPT-6.1 Terrahigh47.3%59.62026-10-26
48Grok 4.6xhigh46.2%56.12026-08-12
49NewGemini 4 Hydrogenlow46.2%–2026-11-12
50Claude Sonnet 5.5low46.0%50.72026-09-28
51Claude Sonnet 5.5medium45.8%50.62026-09-28
52Grok 4.7low45.7%55.42026-09-21
53GPT-6.1 Terralow45.6%56.62026-10-26
54DeepSeek V4 Promax45.2%54.72026-08-13
55DeepSeek V4 Prohigh45.1%56.02026-08-13
56MiniMax M3.1 Flash45.1%54.32026-10-19
57MiniMax M344.9%56.12026-06-01
58Grok 4.7xhigh42.7%57.42026-09-21
59Kimi K3low42.4%53.62026-07-16
60GPT-5.6 Terramax42.3%59.52026-07-09

More knowledge benchmarks · All LLM benchmarks

Cite as: BenchLeader, “AA-Omniscience v1.1: accuracy leaderboard”, data as of 15 Nov 2026, https://www.benchleader.com/benchmarks/aa_omniscience_accuracy.

AA-Omniscience v1.1: accuracy: questions

What does AA-Omniscience v1.1: accuracy measure?
The share of AA-Omniscience questions answered correctly, before any penalty for confident wrong answers. Scores are reported in percent of tasks solved; higher is better.
What is the top score on AA-Omniscience v1.1: accuracy?
As of 15 Nov 2026, the highest score on AA-Omniscience v1.1: accuracy is 79.1% by Claude Fable 5.2 (result dated 9 Nov 2026), ahead of Claude Fable 5.1 at 68.4%. Scores come from Artificial Analysis and are re-read daily.
Which AI model leads AA-Omniscience v1.1: accuracy?
Claude Fable 5.2 leads AA-Omniscience v1.1: accuracy with 79.1% as of 15 Nov 2026 (result dated 9 Nov 2026), ahead of Claude Fable 5.1 at 68.4%.
How many models have AA-Omniscience v1.1: accuracy results?
142 model configurations have a AA-Omniscience v1.1: accuracy result on BenchLeader, all taken from Artificial Analysis.
Who runs AA-Omniscience v1.1: accuracy and how often is it updated?
AA-Omniscience v1.1: accuracy is published by Artificial Analysis. BenchLeader re-reads the published results once a day and records the date each result was published.
Does AA-Omniscience v1.1: accuracy count toward the BenchLeader Index?
No. AA-Omniscience v1.1: accuracy is shown for reference but left out of the composite index.