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.

2026-09-10MODELDeepSeek-V4.1-Flash released with extreme agent memory efficiency and evaluated across VALS • 2026-09-03MODELGPT-6 Astra released by OpenAI - 71.7% on EMB, 63.3% on Tax Agent Bench • 2026-09-02MODELGoogle Gemini 3.8 Flash takes #1 rank on Finance Agent (v2) with 61.4% and 72.2% on EMB • 2026-09-02MODELMeta releases Muse Spark 1.3 & 1.3 Max - 59.0%-60.0% on Finance Agent (v2) • 2026-09-01MODELClaude Fable 5.1 leads Tax Agent Bench (77.6%) and EMB (76.7%) with 75% cheaper cache reads • 2026-09-11BENCHTax Agent Bench and FAB v2 benchmarks updated with latest frontier model rankings • 2026-09-15UPDATEPlatform updated: 43 models tracked across 41 independent benchmarks • 2026-09-10MODELDeepSeek-V4.1-Flash released with extreme agent memory efficiency and evaluated across VALS • 2026-09-03MODELGPT-6 Astra released by OpenAI - 71.7% on EMB, 63.3% on Tax Agent Bench • 2026-09-02MODELGoogle Gemini 3.8 Flash takes #1 rank on Finance Agent (v2) with 61.4% and 72.2% on EMB • 2026-09-02MODELMeta releases Muse Spark 1.3 & 1.3 Max - 59.0%-60.0% on Finance Agent (v2) • 2026-09-01MODELClaude Fable 5.1 leads Tax Agent Bench (77.6%) and EMB (76.7%) with 75% cheaper cache reads • 2026-09-11BENCHTax Agent Bench and FAB v2 benchmarks updated with latest frontier model rankings • 2026-09-15UPDATEPlatform updated: 43 models tracked across 41 independent benchmarks •

Tier 3forensicagentTasks

Tax Agent Bench

Evaluates agents on professional tax compliance, preparation, and advisory reasoning.

Variable tasks

19 models evaluated

Updated 2026

View Source →Submit your score Methodology

77.6%

Top Score

Top model

Claude Fable 5.1

Model Rankings

financebenchmark.ai · Tax Agent

70.8%

🥇

OpenAIGPT-5.560.5%

🥈

AnthropicClaude Opus 575.1%

🥉

AnthropicClaude Sonnet 562.3%

4

OpenAIGPT-5.6 Sol68.0%

5

GoogleGemini 3.7 Flash57.7%

6

xAIGrok 4.670.8%

7

Moonshot AIKimi K3OPEN68.7%

8

MiniMaxMiniMax M3OPEN49.7%

9

DeepSeekDeepSeek V4 Pro (0813)OPEN58.7%

10

OpenAIGPT-5.6 Luna60.8%

11

OpenAIGPT-5.6 Terra65.2%

12

Thinking MachinesInklingOPEN43.5%

13

AlibabaQwen 3.8 MaxOPEN66.0%

14

OpenAIGPT-6 Astra63.3%

15

AnthropicClaude Fable 5.177.6%

16

GoogleGemini 3.8 Flash66.8%

17

MetaMuse Spark 1.371.9%

18

DeepSeekDeepSeek V4.1 FlashOPEN62.5%

19

Zhipu AIGLM 5.3OPEN73.1%

Full Results Table

#	Model	Overall	Source	Date
🥇	Claude Fable 5.1 Anthropic	77.6	VALS (2026)	2026-09
🥈	Claude Opus 5 Anthropic	75.1	VALS (2026)	2026-09
🥉	GLM 5.3OPEN Zhipu AI	73.1	VALS (2026)	2026-09
4	Muse Spark 1.3 Meta	71.9	VALS (2026)	2026-09
5	Grok 4.6 xAI	70.8	VALS (2026)	2026-09
6	Kimi K3OPEN Moonshot AI	68.7	VALS (2026)	2026-09
7	GPT-5.6 Sol OpenAI	68.0	VALS (2026)	2026-09
8	Gemini 3.8 Flash Google	66.8	VALS (2026)	2026-09
9	Qwen 3.8 MaxOPEN Alibaba	66.0	VALS (2026)	2026-09
10	GPT-5.6 Terra OpenAI	65.2	VALS (2026)	2026-09
11	GPT-6 Astra OpenAI	63.3	VALS (2026)	2026-09
12	DeepSeek V4.1 FlashOPEN DeepSeek	62.5	VALS (2026)	2026-09
13	Claude Sonnet 5 Anthropic	62.3	VALS (2026)	2026-09
14	GPT-5.6 Luna OpenAI	60.8	VALS (2026)	2026-09
15	GPT-5.5 OpenAI	60.5	VALS (2026)	2026-09
16	DeepSeek V4 Pro (0813)OPEN DeepSeek	58.7	VALS (2026)	2026-09
17	Gemini 3.7 Flash Google	57.7	VALS (2026)	2026-09
18	MiniMax M3OPEN MiniMax	49.7	VALS (2026)	2026-09
19	InklingOPEN Thinking Machines	43.5	VALS (2026)	2026-09