FinanceBenchmark: Financial AI Leaderboard 2026

FinanceBenchmark provides independent rankings and a composite score for leading Artificial Intelligence models, including Anthropic Claude, OpenAI GPT, and Google Gemini, on complex financial reasoning and forensic accounting tasks. Enable JavaScript to view the interactive, continuously updated leaderboard.

2026-10-20UPDATEPlatform updated: 52 models tracked across 42 independent benchmarks including CorpFin • 2026-10-20BENCHCorpFin added to benchmark suite — long credit agreement comprehension evaluations live • 2026-10-19MODELClaude Sonnet 5.5 and Opus 5.5 achieve 75.9% and 75.8% on Excel Modeling Benchmark v1.1 • 2026-10-19MODELTax Agent Bench updated: Claude Fable 5.1 (49.2%) and Gemini 4 Argon (44.6%) lead corporate tax research • 2026-10-18MODELGemini 4 Argon claims #1 rank on Finance Agent (v2) with 65.4% • 2026-10-13MODELGoogle launches Gemini 4 Argon with 1M-token output limit and native multimodal reasoning • 2026-10-06MODELAnthropic completes Claude 5.5 generation with Claude Haiku 5.5, Opus 5.5, and Sonnet 5.5 • 2026-10-20UPDATEPlatform updated: 52 models tracked across 42 independent benchmarks including CorpFin • 2026-10-20BENCHCorpFin added to benchmark suite — long credit agreement comprehension evaluations live • 2026-10-19MODELClaude Sonnet 5.5 and Opus 5.5 achieve 75.9% and 75.8% on Excel Modeling Benchmark v1.1 • 2026-10-19MODELTax Agent Bench updated: Claude Fable 5.1 (49.2%) and Gemini 4 Argon (44.6%) lead corporate tax research • 2026-10-18MODELGemini 4 Argon claims #1 rank on Finance Agent (v2) with 65.4% • 2026-10-13MODELGoogle launches Gemini 4 Argon with 1M-token output limit and native multimodal reasoning • 2026-10-06MODELAnthropic completes Claude 5.5 generation with Claude Haiku 5.5, Opus 5.5, and Sonnet 5.5 •

Tier 3forensicagentTasks

Tax Agent Bench

Evaluating agents on research-grade US corporate tax compliance and advisory questions.

Variable tasks

25 models evaluated

Updated 2026

View Source →Submit your score Methodology

68.0%

Top Score

Top model

GPT-5.6 Sol

Model Rankings

financebenchmark.ai · Tax Agent

33.1%

🥇

OpenAIGPT-5.560.5%

🥈

AnthropicClaude Opus 545.5%

🥉

AnthropicClaude Sonnet 562.3%

4

OpenAIGPT-5.6 Sol68.0%

5

GoogleGemini 3.7 Flash57.7%

6

xAIGrok 4.633.1%

7

Moonshot AIKimi K3OPEN33.4%

8

MiniMaxMiniMax M3OPEN49.7%

9

AnthropicClaude Fable 538.3%

10

DeepSeekDeepSeek V4 Pro (0813)OPEN58.7%

11

OpenAIGPT-5.6 Luna60.8%

12

OpenAIGPT-5.6 Terra65.2%

13

Thinking MachinesInklingOPEN43.5%

14

MetaMuse Spark 1.232.8%

15

AlibabaQwen 3.8 MaxOPEN66.0%

16

OpenAIGPT-6 Astra63.3%

17

AnthropicClaude Fable 5.149.2%

18

GoogleGemini 3.8 Flash66.8%

19

MetaMuse Spark 1.336.0%

20

MetaMuse Spark 1.3 Max42.4%

21

DeepSeekDeepSeek V4.1 FlashOPEN44.4%

22

Zhipu AIGLM 5.3OPEN39.7%

23

GoogleGemini 4 Argon44.6%

24

AnthropicClaude Opus 5.545.6%

25

AnthropicClaude Sonnet 5.542.0%

Full Results Table

#	Model	Overall	Source	Date
🥇	GPT-5.6 Sol OpenAI	68.0	VALS (2026)	2026-09
🥈	Gemini 3.8 Flash Google	66.8	VALS (2026)	2026-09
🥉	Qwen 3.8 MaxOPEN Alibaba	66.0	VALS (2026)	2026-09
4	GPT-5.6 Terra OpenAI	65.2	VALS (2026)	2026-09
5	GPT-6 Astra OpenAI	63.3	VALS (2026)	2026-09
6	Claude Sonnet 5 Anthropic	62.3	VALS (2026)	2026-09
7	GPT-5.6 Luna OpenAI	60.8	VALS (2026)	2026-09
8	GPT-5.5 OpenAI	60.5	VALS (2026)	2026-09
9	DeepSeek V4 Pro (0813)OPEN DeepSeek	58.7	VALS (2026)	2026-09
10	Gemini 3.7 Flash Google	57.7	VALS (2026)	2026-09
11	MiniMax M3OPEN MiniMax	49.7	VALS (2026)	2026-09
12	Claude Fable 5.1 Anthropic	49.2	VALS (2026)	2026-10
13	Claude Opus 5.5 Anthropic	45.6	VALS (2026)	2026-10
14	Claude Opus 5 Anthropic	45.5	VALS (2026)	2026-10
15	Gemini 4 Argon Google	44.6	VALS (2026)	2026-10
16	DeepSeek V4.1 FlashOPEN DeepSeek	44.4	VALS (2026)	2026-10
17	InklingOPEN Thinking Machines	43.5	VALS (2026)	2026-09
18	Muse Spark 1.3 Max Meta	42.4	VALS (2026)	2026-10
19	Claude Sonnet 5.5 Anthropic	42.0	VALS (2026)	2026-10
20	GLM 5.3OPEN Zhipu AI	39.7	VALS (2026)	2026-10
21	Claude Fable 5 Anthropic	38.3	VALS (2026)	2026-10
22	Muse Spark 1.3 Meta	36.0	VALS (2026)	2026-10
23	Kimi K3OPEN Moonshot AI	33.4	VALS (2026)	2026-10
24	Grok 4.6 xAI	33.1	VALS (2026)	2026-10
25	Muse Spark 1.2 Meta	32.8	VALS (2026)	2026-10

Contents.md

All evaluated models

GPT-5.5

60.5%

Claude Opus 5

45.5%

Claude Sonnet 5

62.3%

GPT-5.6 Sol

68.0%

Gemini 3.7 Flash

57.7%

Grok 4.6

33.1%

Kimi K3

33.4%

MiniMax M3

49.7%

Claude Fable 5

38.3%

DeepSeek V4 Pro (0813)

58.7%

GPT-5.6 Luna

60.8%

GPT-5.6 Terra

65.2%

Inkling

43.5%

Muse Spark 1.2

32.8%

Qwen 3.8 Max

66.0%

GPT-6 Astra

63.3%

Claude Fable 5.1

49.2%

Gemini 3.8 Flash

66.8%

Muse Spark 1.3

36.0%

Muse Spark 1.3 Max

42.4%

DeepSeek V4.1 Flash

44.4%

GLM 5.3

39.7%

Gemini 4 Argon

44.6%

Claude Opus 5.5

45.6%

Claude Sonnet 5.5

42.0%