2026-10-20UPDATEPlatform updated: 52 models tracked across 42 independent benchmarks including CorpFin • 2026-10-20BENCHCorpFin added to benchmark suite — long credit agreement comprehension evaluations live • 2026-10-19MODELClaude Sonnet 5.5 and Opus 5.5 achieve 75.9% and 75.8% on Excel Modeling Benchmark v1.1 • 2026-10-19MODELTax Agent Bench updated: Claude Fable 5.1 (49.2%) and Gemini 4 Argon (44.6%) lead corporate tax research • 2026-10-18MODELGemini 4 Argon claims #1 rank on Finance Agent (v2) with 65.4% • 2026-10-13MODELGoogle launches Gemini 4 Argon with 1M-token output limit and native multimodal reasoning • 2026-10-06MODELAnthropic completes Claude 5.5 generation with Claude Haiku 5.5, Opus 5.5, and Sonnet 5.5 • 2026-10-20UPDATEPlatform updated: 52 models tracked across 42 independent benchmarks including CorpFin • 2026-10-20BENCHCorpFin added to benchmark suite — long credit agreement comprehension evaluations live • 2026-10-19MODELClaude Sonnet 5.5 and Opus 5.5 achieve 75.9% and 75.8% on Excel Modeling Benchmark v1.1 • 2026-10-19MODELTax Agent Bench updated: Claude Fable 5.1 (49.2%) and Gemini 4 Argon (44.6%) lead corporate tax research • 2026-10-18MODELGemini 4 Argon claims #1 rank on Finance Agent (v2) with 65.4% • 2026-10-13MODELGoogle launches Gemini 4 Argon with 1M-token output limit and native multimodal reasoning • 2026-10-06MODELAnthropic completes Claude 5.5 generation with Claude Haiku 5.5, Opus 5.5, and Sonnet 5.5 •
Tier 1agentTasks

Finance Agent Benchmark

Core financial analyst tasks and equity research evaluated using the Finance Agent (FAB v2) harness.

Variable tasks
43 models evaluated
Updated 2026
65.4%
Top Score
Top model
Gemini 4 Argon

Model Rankings

financebenchmark.ai · Finance Agent
Anthropic logo
OpenAI logo
Google logo
DeepSeek logo
xAI logo
Moonshot AI logo
Zhipu AI logo
MiniMax logo
Anthropic logo
Anthropic logo
OpenAI logo
Google logo
DeepSeek logo
xAI logo
Moonshot AI logo
MiniMax logo
Anthropic logo
DeepSeek logo
Google logo
Google logo
Zhipu AI logo
OpenAI logo
OpenAI logo
xAI logo
Thinking Machines logo
Thinking Machines logo
Ant Group logo
Meta logo
Meta logo
NVIDIA logo
Alibaba logo
Alibaba logo
OpenAI logo
Anthropic logo
Google logo
Meta logo
Meta logo
DeepSeek logo
Zhipu AI logo
Zhipu AI logo
Google logo
Anthropic logo
Anthropic logo
44.9%
🥇Anthropic logoAnthropicClaude Opus 4.751.5%
🥈OpenAI logoOpenAIGPT-5.551.8%
🥉Google logoGoogleGemini 3.1 Pro43.0%
4DeepSeek logoDeepSeekDeepSeek V4 ProOPEN44.1%
5xAI logoxAIGrok 4.337.7%
6Moonshot AI logoMoonshot AIKimi K2.6OPEN44.9%
7Zhipu AI logoZhipu AIGLM-5.1OPEN44.8%
8MiniMax logoMiniMaxMiniMax M2.7OPEN27.9%
9Anthropic logoAnthropicClaude Opus 558.6%
10Anthropic logoAnthropicClaude Sonnet 553.9%
11OpenAI logoOpenAIGPT-5.6 Sol53.8%
12Google logoGoogleGemini 3.7 Flash59.0%
13DeepSeek logoDeepSeekDeepSeek V4 FlashOPEN49.5%
14xAI logoxAIGrok 4.653.7%
15Moonshot AI logoMoonshot AIKimi K3OPEN54.4%
16MiniMax logoMiniMaxMiniMax M3OPEN48.3%
17Anthropic logoAnthropicClaude Fable 556.3%
18DeepSeek logoDeepSeekDeepSeek V4 Pro (0813)OPEN50.4%
19Google logoGoogleGemini 3.5 Flash Lite47.4%
20Google logoGoogleGemini 3.6 Flash56.3%
21Zhipu AI logoZhipu AIGLM 5.2OPEN49.7%
22OpenAI logoOpenAIGPT-5.6 Luna55.0%
23OpenAI logoOpenAIGPT-5.6 Terra54.4%
24xAI logoxAIGrok 4.548.4%
25Thinking Machines logoThinking MachinesInklingOPEN46.6%
26Thinking Machines logoThinking MachinesInkling SmallOPEN41.3%
27Ant Group logoAnt GroupLing 3.0 FlashOPEN30.3%
28Meta logoMetaMuse Spark 1.157.2%
29Meta logoMetaMuse Spark 1.260.6%
30NVIDIA logoNVIDIANemotron 3 UltraOPEN37.7%
31Alibaba logoAlibabaQwen 3.7 Plus38.2%
32Alibaba logoAlibabaQwen 3.8 MaxOPEN50.6%
33OpenAI logoOpenAIGPT-6 Astra53.5%
34Anthropic logoAnthropicClaude Fable 5.158.9%
35Google logoGoogleGemini 3.8 Flash61.4%
36Meta logoMetaMuse Spark 1.358.9%
37Meta logoMetaMuse Spark 1.3 Max60.0%
38DeepSeek logoDeepSeekDeepSeek V4.1 FlashOPEN53.5%
39Zhipu AI logoZhipu AIGLM 5.3OPEN55.8%
40Zhipu AI logoZhipu AIGLM 5.3 FlashOPEN57.9%
41Google logoGoogleGemini 4 Argon65.4%
42Anthropic logoAnthropicClaude Opus 5.558.6%
43Anthropic logoAnthropicClaude Sonnet 5.558.1%
Full Results Table
#ModelOverallSourceDate
🥇
Google logoGemini 4 Argon
Google
65.4VALS (2026)2026-10
🥈
Google logoGemini 3.8 Flash
Google
61.4VALS (2026)2026-10
🥉
Meta logoMuse Spark 1.2
Meta
60.6VALS (2026)2026-10
4
Meta logoMuse Spark 1.3 Max
Meta
60.0VALS (2026)2026-10
5
Google logoGemini 3.7 Flash
Google
59.0VALS (2026)2026-10
6
Meta logoMuse Spark 1.3
Meta
58.9VALS (2026)2026-10
7
Anthropic logoClaude Fable 5.1
Anthropic
58.9VALS (2026)2026-10
8
Anthropic logoClaude Opus 5
Anthropic
58.6VALS (2026)2026-10
9
Anthropic logoClaude Opus 5.5
Anthropic
58.6VALS (2026)2026-10
10
Anthropic logoClaude Sonnet 5.5
Anthropic
58.1VALS (2026)2026-10
11
Zhipu AI logoGLM 5.3 FlashOPEN
Zhipu AI
57.9VALS (2026)2026-10
12
Meta logoMuse Spark 1.1
Meta
57.2VALS (2026)2026-10
13
Anthropic logoClaude Fable 5
Anthropic
56.3VALS (2026)2026-09
14
Google logoGemini 3.6 Flash
Google
56.3VALS (2026)2026-09
15
Zhipu AI logoGLM 5.3OPEN
Zhipu AI
55.8VALS (2026)2026-09
16
OpenAI logoGPT-5.6 Luna
OpenAI
55.0VALS (2026)2026-09
17
OpenAI logoGPT-5.6 Terra
OpenAI
54.4VALS (2026)2026-09
18
Moonshot AI logoKimi K3OPEN
Moonshot AI
54.4VALS (2026)2026-09
19
Anthropic logoClaude Sonnet 5
Anthropic
53.9VALS (2026)2026-09
20
OpenAI logoGPT-5.6 Sol
OpenAI
53.8VALS (2026)2026-09
21
xAI logoGrok 4.6
xAI
53.7VALS (2026)2026-09
22
OpenAI logoGPT-6 Astra
OpenAI
53.5VALS (2026)2026-09
23
DeepSeek logoDeepSeek V4.1 FlashOPEN
DeepSeek
53.5VALS (2026)2026-09
24
OpenAI logoGPT-5.5
OpenAI
51.8VALS (2026)2026-09
25
Anthropic logoClaude Opus 4.7
Anthropic
51.5VALS (2026)2026-09
26
Alibaba logoQwen 3.8 MaxOPEN
Alibaba
50.6VALS (2026)2026-09
27
DeepSeek logoDeepSeek V4 Pro (0813)OPEN
DeepSeek
50.4VALS (2026)2026-09
28
Zhipu AI logoGLM 5.2OPEN
Zhipu AI
49.7VALS (2026)2026-09
29
DeepSeek logoDeepSeek V4 FlashOPEN
DeepSeek
49.5VALS (2026)2026-09
30
xAI logoGrok 4.5
xAI
48.4VALS (2026)2026-09
31
MiniMax logoMiniMax M3OPEN
MiniMax
48.3VALS (2026)2026-09
32
Google logoGemini 3.5 Flash Lite
Google
47.4VALS (2026)2026-09
33
Thinking Machines logoInklingOPEN
Thinking Machines
46.6VALS (2026)2026-09
34
Moonshot AI logoKimi K2.6OPEN
Moonshot AI
44.9VALS (2026)2026-09
35
Zhipu AI logoGLM-5.1OPEN
Zhipu AI
44.8VALS (2026)2026-09
36
DeepSeek logoDeepSeek V4 ProOPEN
DeepSeek
44.1VALS (2026)2026-09
37
Google logoGemini 3.1 Pro
Google
43.0VALS (2026)2026-09
38
Thinking Machines logoInkling SmallOPEN
Thinking Machines
41.3VALS (2026)2026-09
39
Alibaba logoQwen 3.7 Plus
Alibaba
38.2VALS (2026)2026-09
40
xAI logoGrok 4.3
xAI
37.7VALS (2026)2026-09
41
NVIDIA logoNemotron 3 UltraOPEN
NVIDIA
37.7VALS (2026)2026-09
42
Ant Group logoLing 3.0 FlashOPEN
Ant Group
30.3VALS (2026)2026-09
43
MiniMax logoMiniMax M2.7OPEN
MiniMax
27.9VALS (2026)2026-09