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 2verificationforensic

TaxEval v2

Comprehensive benchmark for corporate and individual tax code interpretation and compliance reasoning.

Variable tasks

21 models evaluated

Updated 2026

View Source →Submit your score Methodology

80.4%

Top Score

Top model

Muse Spark 1.2

Model Rankings

financebenchmark.ai · TaxEval v2

75.7%

🥇

AnthropicClaude Opus 575.1%

🥈

AnthropicClaude Sonnet 575.6%

🥉

OpenAIGPT-5.6 Sol74.8%

4

DeepSeekDeepSeek V4 FlashOPEN70.7%

5

xAIGrok 4.671.1%

6

Moonshot AIKimi K3OPEN75.7%

7

AnthropicClaude Fable 576.9%

8

DeepSeekDeepSeek V4 Pro (0813)OPEN73.1%

9

GoogleGemini 3.5 Flash Lite72.6%

10

GoogleGemini 3.6 Flash74.9%

11

Zhipu AIGLM 5.2OPEN73.3%

12

OpenAIGPT-5.6 Luna76.2%

13

OpenAIGPT-5.6 Terra76.2%

14

xAIGrok 4.571.7%

15

Thinking MachinesInklingOPEN75.3%

16

Thinking MachinesInkling SmallOPEN75.5%

17

Ant GroupLing 3.0 FlashOPEN70.7%

18

MetaMuse Spark 1.179.7%

19

MetaMuse Spark 1.280.4%

20

NVIDIANemotron 3 UltraOPEN73.1%

21

AlibabaQwen 3.8 MaxOPEN75.5%

Full Results Table

#	Model	Overall	Source	Date
🥇	Muse Spark 1.2 Meta	80.4	VALS (2026)	2026-08
🥈	Muse Spark 1.1 Meta	79.7	VALS (2026)	2026-07
🥉	Claude Fable 5 Anthropic	76.9	VALS (2026)	2026-06
4	GPT-5.6 Luna OpenAI	76.2	VALS (2026)	2026-07
5	GPT-5.6 Terra OpenAI	76.2	VALS (2026)	2026-07
6	Kimi K3OPEN Moonshot AI	75.7	VALS (2026)	2026-07
7	Claude Sonnet 5 Anthropic	75.6	VALS (2026)	2026-06
8	Qwen 3.8 MaxOPEN Alibaba	75.5	VALS (2026)	2026-08
9	Inkling SmallOPEN Thinking Machines	75.5	VALS (2026)	2026-07
10	InklingOPEN Thinking Machines	75.3	VALS (2026)	2026-07
11	Claude Opus 5 Anthropic	75.1	VALS (2026)	2026-07
12	Gemini 3.6 Flash Google	74.9	VALS (2026)	2026-07
13	GPT-5.6 Sol OpenAI	74.8	VALS (2026)	2026-07
14	GLM 5.2OPEN Zhipu AI	73.3	VALS (2026)	2026-06
15	Nemotron 3 UltraOPEN NVIDIA	73.1	VALS (2026)	2026-06
16	DeepSeek V4 Pro (0813)OPEN DeepSeek	73.1	VALS (2026)	2026-08
17	Gemini 3.5 Flash Lite Google	72.6	VALS (2026)	2026-07
18	Grok 4.5 xAI	71.7	VALS (2026)	2026-07
19	Grok 4.6 xAI	71.1	VALS (2026)	2026-08
20	DeepSeek V4 FlashOPEN DeepSeek	70.7	VALS (2026)	2026-07
21	Ling 3.0 FlashOPEN Ant Group	70.7	VALS (2026)	2026-07