EMB v1.1 (Excel financial modeling)

Excel financial models (DCF, LBO, M&A) built by an agent and graded against a rubric. Run by Vals AI.

CategoryAgents
MetricAccuracy (%)
SourceVals AI (evaluator)
LicenseNone stated
Benchmark pagewww.vals.ai
StatusScored
Weight0.8 within Agents
FitDiscrimination 0.55, noise 0.20

Results

#ModelProviderResultDate
1 Claude Fable 5.2 High Anthropic 77.4% 2026-11-09
2 Claude Fable 5.1 Max Anthropic 76.7% 2026-11-09
3 Gemini 4 Carbon High Google 76.5% 2026-11-09
4 Claude Sonnet 5.5 Max Anthropic 75.9% 2026-11-09
5 Claude Opus 5.5 Max Anthropic 75.8% 2026-11-09
6 Muse Wonder Full Max Meta 75.0% 2026-11-09
7 Gemini 4 Argon High Google 74.8% 2026-11-09
8 GPT-6.1 Terra Max OpenAI 73.3% 2026-11-09
9 Gemini 3.8 Flash High Google 72.9% 2026-11-09
10 GPT-6 Astra Max OpenAI 72.7% 2026-11-09
11 Muse Wonder Max Meta 72.5% 2026-11-09
12 Muse Ember Max Meta 72.2% 2026-11-09
13 Grok 4.7 Extra High xAI 71.5% 2026-11-09
14 Kimi K3 Max Moonshot AI 71.3% 2026-11-09
15 GPT-6.1 Sol Max OpenAI 71.1% 2026-11-09
16 Kimi K3.1 Pro Max Moonshot AI 70.0% 2026-11-09
17 GPT-6 Luna Max OpenAI 70.0% 2026-11-09
18 Step 5 Preview StepFun 69.6% 2026-11-09
19 Inkling 2 Small Thinking Machines 69.5% 2026-11-09
20 DeepSeek V4.1 Flash 1017 Max DeepSeek 69.2% 2026-11-09
21 GLM-5.3 Max Zhipu AI 69.1% 2026-11-09
22 Mistral Large 4 High Mistral 69.0% 2026-11-09
23 Hy4 preview Tencent 68.9% 2026-11-09
24 MiMo-V2.6-Flash Xiaomi 68.9% 2026-11-09
25 MiniMax-M3.1-Flash Max MiniMax 67.8% 2026-11-09
26 DeepSeek V4.1 Flash Max DeepSeek 67.8% 2026-11-09
27 MiMo-V2.6-Pro Xiaomi 67.3% 2026-11-09
28 GLM-5.3-Flash Max Zhipu AI 66.3% 2026-11-09
29 Hy4 Tencent 66.2% 2026-11-09
30 Qwen3.8 Max Alibaba 66.0% 2026-11-09
31 Claude Haiku 5.5 Max Anthropic 65.5% 2026-11-09
32 DeepSeek V4 Pro 0813 Max DeepSeek 65.1% 2026-11-09
33 Beam 1.0 Reflection 64.7% 2026-11-09
34 Muse Spark Open Extra High Meta 62.0% 2026-11-09
35 MiniMax-M3 MiniMax 59.8% 2026-11-09
36 Mercury 3 High Inception 57.9% 2026-11-09
37 Nemotron 3.5 Super VL 120B A12B High NVIDIA 52.8% 2026-11-09
38 Inkling Thinking Machines 41.0% 2026-11-09
39 Inkling Small Thinking Machines 40.8% 2026-11-09