BenchLeader
Menu
Best LLMsChoose a modelCoding plansLeaderboardsModelsBenchmarksProvidersReleases7

Analyst Agent (AA) leaderboard

As of 6 Nov 2026, the top score on Analyst Agent (AA) is 60.0% by Gemini 3.7 Flash (Google), ahead of Claude Fable 5.1 at 57.5%, across 52 model configurations with a published result (Artificial Analysis, result dated 13 Aug 2026). The agent is given an analyst brief, a set of documents and tools, and must produce the analysis a junior analyst would. Analyst Agent (AA) is shown for reference only and does not count toward the BenchLeader Index.

Key facts (as of 6 Nov 2026)

  • Gemini 3.7 Flash holds the top Analyst Agent (AA) score at 60.0% (Artificial Analysis, 13 Aug 2026).
  • Claude Fable 5.1 is second on Analyst Agent (AA) at 57.5% (Artificial Analysis, 1 Sep 2026).
  • Analyst Agent (AA): Multi-step financial-analyst tasks with tools, run by Artificial Analysis (Artificial Analysis).
  • 52 model configurations have a published Analyst Agent (AA) result on BenchLeader (BenchLeader, 6 Nov 2026).
  • Analyst Agent (AA) does not feed the BenchLeader Index; it is shown for reference (BenchLeader methodology).

Multi-step financial-analyst tasks with tools, run by Artificial Analysis.

Published by
Artificial Analysis
Category
Agents & tools
Index weight
Reference only
Models
52
Data as of
6 Nov 2026

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

What the test looks like

The agent is given an analyst brief, a set of documents and tools, and must produce the analysis a junior analyst would.

How it is scored

Rubric score across tasks, run by Artificial Analysis.

What to keep in mind

New, with a small number of models measured so far.

Top 15
  1. 1Gemini 3.7 Flash (high)60.0%
  2. 2Claude Fable 5.1 (max)57.5%
  3. 3Claude Sonnet 5.5 (max)57.5%
  4. 4Claude Opus 5.5 (max)56.3%
  5. 5Grok 4.7 (high)55.0%
  6. 6Claude Opus 5 (max)53.8%
  7. 7Claude Haiku 5.5 (max)53.8%
  8. 8Grok 4.7 (xhigh)52.5%
  9. 9GPT-6 Astra (max)51.3%
  10. 10Inkling 2 Small51.3%
  11. 11GPT-6.1 Sol (max)50.0%
  12. 12GPT-5.5 (xhigh)50.0%
  13. 13Gemini 4 Argon (high)48.8%
  14. 14Claude Fable 5 (max)48.8%
  15. 15GPT-5.6 Sol (max)47.5%
52 of 52
#
1Gemini 3.7 Flashhigh60.0%62.22026-08-13
2Claude Fable 5.1max57.5%67.22026-09-01
3Claude Sonnet 5.5max57.5%65.42026-09-28
4Claude Opus 5.5max56.3%69.32026-09-22
5Grok 4.7high55.0%57.72026-09-21
6Claude Opus 5max53.8%65.42026-07-24
7Claude Haiku 5.5max53.8%57.02026-10-07
8Grok 4.7xhigh52.5%58.32026-09-21
9GPT-6 Astramax51.3%67.82026-09-03
10NewInkling 2 Small51.3%57.42026-11-05
11GPT-6.1 Solmax50.0%67.32026-09-29
12GPT-5.5xhigh50.0%64.32026-04-23
13Gemini 4 Argonhigh48.8%66.02026-09-30
14Claude Fable 5max48.8%65.32026-06-09
15GPT-5.6 Solmax47.5%65.42026-07-09
16Gemini 3.8 Flashhigh46.3%62.02026-09-02
17Claude Sonnet 5max46.3%53.82026-06-30
18Gemini 3.5 Flashhigh45.0%62.52026-05-19
19Claude Opus 4.8max45.0%60.22026-05-28
20Qwen3.8 Maxmax45.0%58.82026-09-02
21Claude Opus 4.7max43.8%59.52026-04-16
22GPT-6 Lunamax43.8%56.72026-09-22
23Gemini 3.1 Pro41.3%60.52026-02-19
24Grok 4.6high41.3%60.52026-08-12
25Kimi K3max38.8%62.12026-07-16
26Kimi K3.1 Pro38.8%61.72026-10-23
27GLM 5.3max38.8%59.02026-08-18
28Muse Spark 1.3max37.5%62.02026-09-02
29MiMo-V2.6-Pro37.5%57.32026-09-21
30DeepSeek V4.1 Flashmax35.0%59.02026-09-10
31Grok 4.5high35.0%56.92026-07-08
32Step 5 Preview35.0%54.92026-09-18
33Inkling Small27.5%52.12026-07-30
34GLM 5.3 Flash25.0%56.82026-08-26
35DeepSeek V4 Flash 0420max25.0%–2026-04-24
36Inklingxhigh23.8%50.22026-07-15
37Claude Sonnet 4.6max20.0%54.22026-02-17
38Muse Glimmerhigh20.0%46.62026-08-10
39MiMo-V2.5-Prothinking20.0%–2026-04-22
40Qwen3.7max18.8%60.32026-05-19
41DeepSeek V4 Pro 0424max18.8%–2026-04-24
42Ling 3.0 Flash Fin16.3%–2026-09-11
43GPT-5.4 Minixhigh15.0%56.02026-03-17
44Claude Haiku 4.5thinking15.0%45.42025-10-15
45Mistral Medium 3.512.5%46.62026-04-29
46MiniMax M2.711.3%56.52026-03-18
47MiniMax M310.0%56.42026-06-01
48Grok 4.3high8.8%56.62026-04-30
49Gemini 3.1 Flash Lite8.8%52.42026-03-03
50GPT-5.5no reasoning7.5%52.32026-04-23
51Nemotron 3 Ultra 550B A55Bthinking6.3%49.42026-06-04
52Mistral Small 4thinking1.3%40.52026-03-16

More agents & tools benchmarks · All LLM benchmarks

Related benchmarks

Top head-to-heads on Analyst Agent (AA)

Cite as: BenchLeader, “Analyst Agent (AA) leaderboard”, data as of 6 Nov 2026, https://www.benchleader.com/benchmarks/aa_analyst_agent.

Analyst Agent (AA): questions

What does Analyst Agent (AA) measure?
The agent is given an analyst brief, a set of documents and tools, and must produce the analysis a junior analyst would. Scores are reported in percent of tasks solved; higher is better.
What is the top score on Analyst Agent (AA)?
As of 6 Nov 2026, the highest score on Analyst Agent (AA) is 60.0% by Gemini 3.7 Flash (result dated 13 Aug 2026), ahead of Claude Fable 5.1 at 57.5%. Scores come from Artificial Analysis and are re-read daily.
Which AI model leads Analyst Agent (AA)?
Gemini 3.7 Flash leads Analyst Agent (AA) with 60.0% as of 6 Nov 2026 (result dated 13 Aug 2026), ahead of Claude Fable 5.1 at 57.5%.
How many models have Analyst Agent (AA) results?
52 model configurations have a Analyst Agent (AA) result on BenchLeader, all taken from Artificial Analysis.
Who runs Analyst Agent (AA) and how often is it updated?
Analyst Agent (AA) is published by Artificial Analysis. BenchLeader re-reads the published results once a day and records the date each result was published.
Does Analyst Agent (AA) count toward the BenchLeader Index?
No. Analyst Agent (AA) is shown for reference but left out of the composite index.