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 3agentTasks

ProgramBench

Program synthesis and execution verification across complex algorithmic challenges.

Variable tasks

17 models evaluated

Updated 2026

View Source →Submit your score Methodology

3.0%

Top Score

Top model

Claude Opus 5

Model Rankings

financebenchmark.ai · ProgramBench

0.0%

🥇

AnthropicClaude Opus 53.0%

🥈

AnthropicClaude Sonnet 50.0%

🥉

OpenAIGPT-5.6 Sol1.5%

4

Moonshot AIKimi K3OPEN2.0%

5

AnthropicClaude Fable 52.0%

6

GoogleGemini 3.5 Flash Lite0.0%

7

GoogleGemini 3.6 Flash0.0%

8

Zhipu AIGLM 5.2OPEN0.5%

9

OpenAIGPT-5.6 Luna0.0%

10

OpenAIGPT-5.6 Terra0.5%

11

xAIGrok 4.50.0%

12

Thinking MachinesInklingOPEN0.0%

13

Thinking MachinesInkling SmallOPEN0.5%

14

Moonshot AIKimi K2.7 CodeOPEN0.0%

15

MetaMuse Spark 1.10.0%

16

NVIDIANemotron 3 UltraOPEN0.0%

17

AlibabaQwen 3.8 MaxOPEN0.0%

Full Results Table

#	Model	Overall	Source	Date
🥇	Claude Opus 5 Anthropic	3.0	VALS (2026)	2026-07
🥈	Kimi K3OPEN Moonshot AI	2.0	VALS (2026)	2026-07
🥉	Claude Fable 5 Anthropic	2.0	VALS (2026)	2026-06
4	GPT-5.6 Sol OpenAI	1.5	VALS (2026)	2026-07
5	GLM 5.2OPEN Zhipu AI	0.5	VALS (2026)	2026-06
6	GPT-5.6 Terra OpenAI	0.5	VALS (2026)	2026-07
7	Inkling SmallOPEN Thinking Machines	0.5	VALS (2026)	2026-07
8	Claude Sonnet 5 Anthropic	0.0	VALS (2026)	2026-06
9	Gemini 3.5 Flash Lite Google	0.0	VALS (2026)	2026-07
10	Gemini 3.6 Flash Google	0.0	VALS (2026)	2026-07
11	GPT-5.6 Luna OpenAI	0.0	VALS (2026)	2026-07
12	Grok 4.5 xAI	0.0	VALS (2026)	2026-07
13	InklingOPEN Thinking Machines	0.0	VALS (2026)	2026-07
14	Kimi K2.7 CodeOPEN Moonshot AI	0.0	VALS (2026)	2026-06
15	Muse Spark 1.1 Meta	0.0	VALS (2026)	2026-07
16	Nemotron 3 UltraOPEN NVIDIA	0.0	VALS (2026)	2026-06
17	Qwen 3.8 MaxOPEN Alibaba	0.0	VALS (2026)	2026-08