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 3documentQAforensic

Legal Research Bench

Evaluates statutory interpretation, precedent discovery, and legal brief synthesis.

Variable tasks

22 models evaluated

Updated 2026

View Source →Submit your score Methodology

55.3%

Top Score

Top model

Claude Opus 5

Model Rankings

financebenchmark.ai · Legal Research

44.2%

🥇

AnthropicClaude Opus 555.3%

🥈

AnthropicClaude Sonnet 541.8%

🥉

OpenAIGPT-5.6 Sol48.1%

4

DeepSeekDeepSeek V4 FlashOPEN30.3%

5

xAIGrok 4.648.1%

6

Moonshot AIKimi K3OPEN44.2%

7

AnthropicClaude Fable 549.5%

8

DeepSeekDeepSeek V4 Pro (0813)OPEN40.9%

9

GoogleGemini 3.5 Flash Lite13.9%

10

GoogleGemini 3.6 Flash25.0%

11

Zhipu AIGLM 5.2OPEN31.3%

12

OpenAIGPT-5.6 Luna36.5%

13

OpenAIGPT-5.6 Terra40.9%

14

xAIGrok 4.538.0%

15

Thinking MachinesInklingOPEN28.4%

16

Thinking MachinesInkling SmallOPEN25.5%

17

Ant GroupLing 3.0 FlashOPEN0.0%

18

MetaMuse Spark 1.138.0%

19

MetaMuse Spark 1.243.8%

20

NVIDIANemotron 3 UltraOPEN15.4%

21

AlibabaQwen 3.7 Plus16.4%

22

AlibabaQwen 3.8 MaxOPEN47.6%

Full Results Table

#	Model	Overall	Source	Date
🥇	Claude Opus 5 Anthropic	55.3	VALS (2026)	2026-07
🥈	Claude Fable 5 Anthropic	49.5	VALS (2026)	2026-06
🥉	GPT-5.6 Sol OpenAI	48.1	VALS (2026)	2026-07
4	Grok 4.6 xAI	48.1	VALS (2026)	2026-08
5	Qwen 3.8 MaxOPEN Alibaba	47.6	VALS (2026)	2026-08
6	Kimi K3OPEN Moonshot AI	44.2	VALS (2026)	2026-07
7	Muse Spark 1.2 Meta	43.8	VALS (2026)	2026-08
8	Claude Sonnet 5 Anthropic	41.8	VALS (2026)	2026-06
9	DeepSeek V4 Pro (0813)OPEN DeepSeek	40.9	VALS (2026)	2026-08
10	GPT-5.6 Terra OpenAI	40.9	VALS (2026)	2026-07
11	Grok 4.5 xAI	38.0	VALS (2026)	2026-07
12	Muse Spark 1.1 Meta	38.0	VALS (2026)	2026-07
13	GPT-5.6 Luna OpenAI	36.5	VALS (2026)	2026-07
14	GLM 5.2OPEN Zhipu AI	31.3	VALS (2026)	2026-06
15	DeepSeek V4 FlashOPEN DeepSeek	30.3	VALS (2026)	2026-07
16	InklingOPEN Thinking Machines	28.4	VALS (2026)	2026-07
17	Inkling SmallOPEN Thinking Machines	25.5	VALS (2026)	2026-07
18	Gemini 3.6 Flash Google	25.0	VALS (2026)	2026-07
19	Qwen 3.7 Plus Alibaba	16.4	VALS (2026)	2026-06
20	Nemotron 3 UltraOPEN NVIDIA	15.4	VALS (2026)	2026-06
21	Gemini 3.5 Flash Lite Google	13.9	VALS (2026)	2026-07
22	Ling 3.0 FlashOPEN Ant Group	0.0	VALS (2026)	2026-07