BenchLeader
Menu
Best LLMsChoose a modelCoding plansLeaderboardsModelsBenchmarksProvidersReleases12

AA-Briefcase leaderboard

As of 9 Nov 2026, the top score on AA-Briefcase is 1813 by Claude Sonnet 5.5 (Anthropic), ahead of Claude Opus 5.5 at 1800, across 167 model configurations with a published result (Artificial Analysis, result dated 28 Sep 2026). Knowledge work delivered as real files — spreadsheets, decks, documents — graded head to head, so the score is an Elo rather than a percentage. AA-Briefcase is shown for reference only and does not count toward the BenchLeader Index.

Key facts (as of 9 Nov 2026)

  • Claude Sonnet 5.5 holds the top AA-Briefcase v1.1 score at 1813 (Artificial Analysis, 28 Sep 2026).
  • Claude Opus 5.5 is second on AA-Briefcase v1.1 at 1800 (Artificial Analysis, 22 Sep 2026).
  • AA-Briefcase v1.1: Knowledge work delivered as real files — spreadsheets, decks, documents — graded head to head, so the score is an Elo rather than a percentage (Artificial Analysis).
  • 167 model configurations have a published AA-Briefcase v1.1 result on BenchLeader (BenchLeader, 9 Nov 2026).
  • AA-Briefcase v1.1 does not feed the BenchLeader Index; it is shown for reference (BenchLeader methodology).

Knowledge work delivered as real files — spreadsheets, decks, documents — graded head to head, so the score is an Elo rather than a percentage.

Published by
Artificial Analysis
Category
Agents & tools
Index weight
Reference only
Models
167
Data as of
9 Nov 2026

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

Top 15
  1. 1Claude Sonnet 5.5 (max)1813
  2. 2Claude Opus 5.5 (max)1800
  3. 3Claude Fable 5.2 (max)1785
  4. 4Claude Opus 5.5 (xhigh)1760
  5. 5Claude Sonnet 5.5 (xhigh)1742
  6. 6Claude Fable 5.2 (xhigh)1738
  7. 7Muse Wonder Full (max)1729
  8. 8Claude Opus 5.5 (high)1682
  9. 9Claude Fable 5.2 (high)1677
  10. 10Kimi K3.1 Pro1674
  11. 11Claude Fable 5.1 (max)1669
  12. 12Muse Wonder (max)1659
  13. 13Claude Opus 5 (max)1655
  14. 14Grok 4.7 (xhigh)1637
  15. 15Claude Sonnet 5.5 (high)1634
167 of 167
#
1Claude Sonnet 5.5max181365.0$20.47625k2026-09-28
2Claude Opus 5.5max180068.9$21.05338k2026-09-22
3NewClaude Fable 5.2max178569.4$59.35550k2026-11-09
4Claude Opus 5.5xhigh176065.5$12.27198k2026-09-22
5Claude Sonnet 5.5xhigh174262.7$6.98242k2026-09-28
6NewClaude Fable 5.2xhigh173868.7$28.56287k2026-11-09
7Muse Wonder Fullmax172962.2–107k2026-10-12
8Claude Opus 5.5high168265.4$6.27113k2026-09-22
9NewClaude Fable 5.2high167768.9$15.49168k2026-11-09
10Kimi K3.1 Pro167461.2$7.22144k2026-10-23
11Claude Fable 5.1max166966.9$22.71187k2026-09-01
12Muse Wondermax165961.9$8.68112k2026-10-12
13Claude Opus 5max165565.1$17.79170k2026-07-24
14Grok 4.7xhigh163758.1$8.03161k2026-09-21
15Claude Sonnet 5.5high163458.0$3.27140k2026-09-28
16Grok 4.7high162957.0$6.38137k2026-09-21
17NewClaude Fable 5.2medium162266.3$6.8480k2026-11-09
18Claude Opus 5.5medium162061.5$4.4083k2026-09-22
19Gemini 4 Carbonhigh161168.5$8.16138k2026-10-28
20Muse Wonderhigh161056.3$5.6782k2026-10-12
21Qwen3.8 Maxmax160658.9$19.58334k2026-09-02
22Gemini 4 Carbonmedium160264.1$7.06122k2026-10-28
23Muse Spark 1.3max158161.7$3.28105k2026-09-02
24Claude Haiku 5.5max157456.8$0.675425k2026-10-07
25Qwen3.8-Flash-Next157454.4––2026-08-26
26DeepSeek V4.1 Flash 1017max157058.9$0.678202k2026-10-17
27GPT-6 Astramax156967.4$9.5069k2026-09-03
28Gemini 4 Carbonlow1567–$7.2087k2026-10-28
29NewClaude Fable 5.2low156365.3$5.6267k2026-11-09
30GPT-6.1 Solmax155567.0$2.31115k2026-09-29
31GPT-6 Astraxhigh154466.2$6.5645k2026-09-03
32Claude Fable 5max153565.3––2026-06-09
33Grok 4.6high152860.5––2026-08-12
34NewMuse Emberxhigh1527–$0.10274k2026-11-09
35NewInkling 2 Small152758.3$1.02162k2026-11-05
36Claude Haiku 5.5xhigh152356.8$0.366240k2026-10-07
37Ling 3.1 Flash152250.8$3.88350k2026-10-22
38NewMuse Embermax1521–$0.13591k2026-11-09
39MiMo-V2.6-Pro151257.1$0.380164k2026-09-21
40GPT-6 Astrahigh150566.5$4.7632k2026-09-03
41GPT-6.1 Solxhigh150363.8$1.0444k2026-09-29
42GLM 5.3max150358.9$5.65136k2026-08-18
43Kimi K3max149762.0––2026-07-16
44NewAgnes 2.5 Pro Theta149650.8$1.34281k2026-11-07
45Grok 4.7low149355.5$2.1069k2026-09-21
46MiMo-V2.6-Flash148754.9$0.152161k2026-09-21
47Gemini 4 Argonhigh148565.7$11.66142k2026-09-30
48GPT-5.6 Solmax147965.4––2026-07-09
49GPT-6 Solmax147662.9$2.6268k2026-09-22
50MiniMax M3.1 Flash147653.9$0.529155k2026-10-19
51GPT-6.1 Solhigh146963.5$0.83734k2026-09-29
52GPT-6 Astramedium146164.5$3.9426k2026-09-03
53GLM 5.3 Flash145156.8$0.984132k2026-08-26
54Hy4144755.3$1.3189k2026-11-03
55Claude Haiku 5.5high144155.5$0.223157k2026-10-07
56Claude Sonnet 5.5medium143950.9$1.3972k2026-09-28
57Qwen3.8 2.4T A95B143051.6––2026-08-12
58DeepSeek V4.1 Flash 1017no reasoning142545.0$0.816121k2026-10-17
59Step 5 Preview142554.6$2.37120k2026-09-18
60Deepseek v4 Flash Visionmax1418–––2026-08-21

More agents & tools benchmarks · All LLM benchmarks

Related benchmarks

Top head-to-heads on AA-Briefcase v1.1

Cite as: BenchLeader, “AA-Briefcase v1.1 leaderboard”, data as of 9 Nov 2026, https://www.benchleader.com/benchmarks/aa_briefcase.

AA-Briefcase v1.1: questions

What does AA-Briefcase v1.1 measure?
Knowledge work delivered as real files — spreadsheets, decks, documents — graded head to head, so the score is an Elo rather than a percentage. Scores are reported in Elo-style ratings from pairwise comparisons; higher is better.
What is the top score on AA-Briefcase?
As of 9 Nov 2026, the highest score on AA-Briefcase v1.1 is 1813 by Claude Sonnet 5.5 (result dated 28 Sep 2026), ahead of Claude Opus 5.5 at 1800. Scores come from Artificial Analysis and are re-read daily.
Which AI model leads AA-Briefcase v1.1?
Claude Sonnet 5.5 leads AA-Briefcase v1.1 with 1813 as of 9 Nov 2026 (result dated 28 Sep 2026), ahead of Claude Opus 5.5 at 1800.
How many models have AA-Briefcase v1.1 results?
167 model configurations have a AA-Briefcase v1.1 result on BenchLeader, all taken from Artificial Analysis.
Who runs AA-Briefcase v1.1 and how often is it updated?
AA-Briefcase v1.1 is published by Artificial Analysis. BenchLeader re-reads the published results once a day and records the date each result was published.
Does AA-Briefcase v1.1 count toward the BenchLeader Index?
No. AA-Briefcase v1.1 is shown for reference but left out of the composite index.