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 3documentQA

MedScribe

Clinical consultation documentation, transcription accuracy, and medical reporting.

Variable tasks

20 models evaluated

Updated 2026

View Source →Submit your score Methodology

91.0%

Top Score

Top model

Claude Opus 5

Model Rankings

financebenchmark.ai · MedScribe

88.0%

🥇

AnthropicClaude Opus 591.0%

🥈

AnthropicClaude Sonnet 576.0%

🥉

OpenAIGPT-5.6 Sol85.2%

4

DeepSeekDeepSeek V4 FlashOPEN80.4%

5

xAIGrok 4.686.5%

6

Moonshot AIKimi K3OPEN88.0%

7

AnthropicClaude Fable 588.5%

8

DeepSeekDeepSeek V4 Pro (0813)OPEN80.2%

9

GoogleGemini 3.5 Flash Lite70.9%

10

GoogleGemini 3.6 Flash79.7%

11

Zhipu AIGLM 5.2OPEN83.5%

12

OpenAIGPT-5.6 Luna84.4%

13

OpenAIGPT-5.6 Terra82.9%

14

xAIGrok 4.586.9%

15

Thinking MachinesInklingOPEN85.4%

16

Thinking MachinesInkling SmallOPEN84.1%

17

Ant GroupLing 3.0 FlashOPEN80.9%

18

MetaMuse Spark 1.188.9%

19

MetaMuse Spark 1.290.1%

20

AlibabaQwen 3.8 MaxOPEN85.0%

Full Results Table

#	Model	Overall	Source	Date
🥇	Claude Opus 5 Anthropic	91.0	VALS (2026)	2026-07
🥈	Muse Spark 1.2 Meta	90.1	VALS (2026)	2026-08
🥉	Muse Spark 1.1 Meta	88.9	VALS (2026)	2026-07
4	Claude Fable 5 Anthropic	88.5	VALS (2026)	2026-06
5	Kimi K3OPEN Moonshot AI	88.0	VALS (2026)	2026-07
6	Grok 4.5 xAI	86.9	VALS (2026)	2026-07
7	Grok 4.6 xAI	86.5	VALS (2026)	2026-08
8	InklingOPEN Thinking Machines	85.4	VALS (2026)	2026-07
9	GPT-5.6 Sol OpenAI	85.2	VALS (2026)	2026-07
10	Qwen 3.8 MaxOPEN Alibaba	85.0	VALS (2026)	2026-08
11	GPT-5.6 Luna OpenAI	84.4	VALS (2026)	2026-07
12	Inkling SmallOPEN Thinking Machines	84.1	VALS (2026)	2026-07
13	GLM 5.2OPEN Zhipu AI	83.5	VALS (2026)	2026-06
14	GPT-5.6 Terra OpenAI	82.9	VALS (2026)	2026-07
15	Ling 3.0 FlashOPEN Ant Group	80.9	VALS (2026)	2026-07
16	DeepSeek V4 FlashOPEN DeepSeek	80.4	VALS (2026)	2026-07
17	DeepSeek V4 Pro (0813)OPEN DeepSeek	80.2	VALS (2026)	2026-08
18	Gemini 3.6 Flash Google	79.7	VALS (2026)	2026-07
19	Claude Sonnet 5 Anthropic	76.0	VALS (2026)	2026-06
20	Gemini 3.5 Flash Lite Google	70.9	VALS (2026)	2026-07