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 •
Home/Benchmarks/EMB
Tier 3documentQAforensic

EMB

Evaluates multi-step enterprise business operations, accounting workflows, and strategic synthesis.

Variable tasks
39 models evaluated
Updated 2026
VALS (2026)
View Source →Submit your scoreMethodology
76.7%
Top Score
Top model
Claude Fable 5.1
Model Rankings
financebenchmark.ai · EMB
Anthropic
OpenAI
Google
DeepSeek
xAI
Moonshot AI
MiniMax
Anthropic
Anthropic
OpenAI
Google
DeepSeek
xAI
Moonshot AI
MiniMax
Anthropic
DeepSeek
Google
Google
Zhipu AI
OpenAI
OpenAI
xAI
Thinking Machines
Thinking Machines
Ant Group
Meta
Meta
NVIDIA
Alibaba
Alibaba
OpenAI
Anthropic
Google
Meta
Meta
DeepSeek
Zhipu AI
Zhipu AI
57.9%
🥇AnthropicAnthropicClaude Opus 4.763.7%
🥈OpenAIOpenAIGPT-5.564.5%
🥉GoogleGoogleGemini 3.1 Pro52.6%
4DeepSeekDeepSeekDeepSeek V4 ProOPEN51.6%
5xAIxAIGrok 4.318.4%
6Moonshot AIMoonshot AIKimi K2.6OPEN57.9%
7MiniMaxMiniMaxMiniMax M2.7OPEN28.4%
8AnthropicAnthropicClaude Opus 573.6%
9AnthropicAnthropicClaude Sonnet 566.3%
10OpenAIOpenAIGPT-5.6 Sol72.3%
11GoogleGoogleGemini 3.7 Flash71.3%
12DeepSeekDeepSeekDeepSeek V4 FlashOPEN57.0%
13xAIxAIGrok 4.662.7%
14Moonshot AIMoonshot AIKimi K3OPEN66.4%
15MiniMaxMiniMaxMiniMax M3OPEN47.8%
16AnthropicAnthropicClaude Fable 573.7%
17DeepSeekDeepSeekDeepSeek V4 Pro (0813)OPEN52.8%
18GoogleGoogleGemini 3.5 Flash Lite43.2%
19GoogleGoogleGemini 3.6 Flash65.4%
20Zhipu AIZhipu AIGLM 5.2OPEN61.5%
21OpenAIOpenAIGPT-5.6 Luna67.1%
22OpenAIOpenAIGPT-5.6 Terra66.2%
23xAIxAIGrok 4.552.9%
24Thinking MachinesThinking MachinesInklingOPEN38.4%
25Thinking MachinesThinking MachinesInkling SmallOPEN31.8%
26Ant GroupAnt GroupLing 3.0 FlashOPEN25.0%
27MetaMetaMuse Spark 1.156.4%
28MetaMetaMuse Spark 1.257.0%
29NVIDIANVIDIANemotron 3 UltraOPEN31.7%
30AlibabaAlibabaQwen 3.7 Plus49.4%
31AlibabaAlibabaQwen 3.8 MaxOPEN60.1%
32OpenAIOpenAIGPT-6 Astra71.7%
33AnthropicAnthropicClaude Fable 5.176.7%
34GoogleGoogleGemini 3.8 Flash72.2%
35MetaMetaMuse Spark 1.362.7%
36MetaMetaMuse Spark 1.3 Max67.4%
37DeepSeekDeepSeekDeepSeek V4.1 FlashOPEN57.2%
38Zhipu AIZhipu AIGLM 5.3OPEN56.3%
39Zhipu AIZhipu AIGLM 5.3 FlashOPEN55.9%
Full Results Table
#ModelOverallSourceDate
🥇
AnthropicClaude Fable 5.1
Anthropic
76.7VALS (2026)2026-09
🥈
AnthropicClaude Fable 5
Anthropic
73.7VALS (2026)2026-09
🥉
AnthropicClaude Opus 5
Anthropic
73.6VALS (2026)2026-09
4
OpenAIGPT-5.6 Sol
OpenAI
72.3VALS (2026)2026-09
5
GoogleGemini 3.8 Flash
Google
72.2VALS (2026)2026-09
6
OpenAIGPT-6 Astra
OpenAI
71.7VALS (2026)2026-09
7
GoogleGemini 3.7 Flash
Google
71.3VALS (2026)2026-09
8
MetaMuse Spark 1.3 Max
Meta
67.4VALS (2026)2026-09
9
OpenAIGPT-5.6 Luna
OpenAI
67.1VALS (2026)2026-09
10
Moonshot AIKimi K3OPEN
Moonshot AI
66.4VALS (2026)2026-09
11
AnthropicClaude Sonnet 5
Anthropic
66.3VALS (2026)2026-09
12
OpenAIGPT-5.6 Terra
OpenAI
66.2VALS (2026)2026-09
13
GoogleGemini 3.6 Flash
Google
65.4VALS (2026)2026-09
14
OpenAIGPT-5.5
OpenAI
64.5VALS (2026)2026-09
15
AnthropicClaude Opus 4.7
Anthropic
63.7VALS (2026)2026-09
16
xAIGrok 4.6
xAI
62.7VALS (2026)2026-09
17
MetaMuse Spark 1.3
Meta
62.7VALS (2026)2026-09
18
Zhipu AIGLM 5.2OPEN
Zhipu AI
61.5VALS (2026)2026-09
19
AlibabaQwen 3.8 MaxOPEN
Alibaba
60.1VALS (2026)2026-09
20
Moonshot AIKimi K2.6OPEN
Moonshot AI
57.9VALS (2026)2026-09
21
DeepSeekDeepSeek V4.1 FlashOPEN
DeepSeek
57.2VALS (2026)2026-09
22
DeepSeekDeepSeek V4 FlashOPEN
DeepSeek
57.0VALS (2026)2026-09
23
MetaMuse Spark 1.2
Meta
57.0VALS (2026)2026-09
24
MetaMuse Spark 1.1
Meta
56.4VALS (2026)2026-09
25
Zhipu AIGLM 5.3OPEN
Zhipu AI
56.3VALS (2026)2026-09
26
Zhipu AIGLM 5.3 FlashOPEN
Zhipu AI
55.9VALS (2026)2026-09
27
xAIGrok 4.5
xAI
52.9VALS (2026)2026-09
28
DeepSeekDeepSeek V4 Pro (0813)OPEN
DeepSeek
52.8VALS (2026)2026-09
29
GoogleGemini 3.1 Pro
Google
52.6VALS (2026)2026-09
30
DeepSeekDeepSeek V4 ProOPEN
DeepSeek
51.6VALS (2026)2026-09
31
AlibabaQwen 3.7 Plus
Alibaba
49.4VALS (2026)2026-09
32
MiniMaxMiniMax M3OPEN
MiniMax
47.8VALS (2026)2026-09
33
GoogleGemini 3.5 Flash Lite
Google
43.2VALS (2026)2026-09
34
Thinking MachinesInklingOPEN
Thinking Machines
38.4VALS (2026)2026-09
35
Thinking MachinesInkling SmallOPEN
Thinking Machines
31.8VALS (2026)2026-09
36
NVIDIANemotron 3 UltraOPEN
NVIDIA
31.7VALS (2026)2026-09
37
MiniMaxMiniMax M2.7OPEN
MiniMax
28.4VALS (2026)2026-09
38
Ant GroupLing 3.0 FlashOPEN
Ant Group
25.0VALS (2026)2026-09
39
xAIGrok 4.3
xAI
18.4VALS (2026)2026-09
All evaluated models
Anthropic
Claude Opus 4.7
63.7%
OpenAI
GPT-5.5
64.5%
Google
Gemini 3.1 Pro
52.6%
DeepSeek
DeepSeek V4 Pro
51.6%
xAI
Grok 4.3
18.4%
Moonshot AI
Kimi K2.6
57.9%
MiniMax
MiniMax M2.7
28.4%
Anthropic
Claude Opus 5
73.6%
Anthropic
Claude Sonnet 5
66.3%
OpenAI
GPT-5.6 Sol
72.3%
Google
Gemini 3.7 Flash
71.3%
DeepSeek
DeepSeek V4 Flash
57.0%
xAI
Grok 4.6
62.7%
Moonshot AI
Kimi K3
66.4%
MiniMax
MiniMax M3
47.8%
Anthropic
Claude Fable 5
73.7%
DeepSeek
DeepSeek V4 Pro (0813)
52.8%
Google
Gemini 3.5 Flash Lite
43.2%
Google
Gemini 3.6 Flash
65.4%
Zhipu AI
GLM 5.2
61.5%
OpenAI
GPT-5.6 Luna
67.1%
OpenAI
GPT-5.6 Terra
66.2%
xAI
Grok 4.5
52.9%
Thinking Machines
Inkling
38.4%
Thinking Machines
Inkling Small
31.8%
Ant Group
Ling 3.0 Flash
25.0%
Meta
Muse Spark 1.1
56.4%
Meta
Muse Spark 1.2
57.0%
NVIDIA
Nemotron 3 Ultra
31.7%
Alibaba
Qwen 3.7 Plus
49.4%
Alibaba
Qwen 3.8 Max
60.1%
OpenAI
GPT-6 Astra
71.7%
Anthropic
Claude Fable 5.1
76.7%
Google
Gemini 3.8 Flash
72.2%
Meta
Muse Spark 1.3
62.7%
Meta
Muse Spark 1.3 Max
67.4%
DeepSeek
DeepSeek V4.1 Flash
57.2%
Zhipu AI
GLM 5.3
56.3%
Zhipu AI
GLM 5.3 Flash
55.9%