Skip to content

Benchmark

FinBen FinArg-ARC MicroF1

Scores are percentages. Higher is better.

Models measured
19
Scores recorded
19
Sources
1
IQ weight
0.38

Score distribution

51.6%
Mean
53.1%
Median
10.4%
Std dev
39%
10th percentile
63%
90th percentile
27%
Min
27%66%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.38 for chat models, measured over 19 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Mistral (7B) Instruct v0.3published-demo-leaderboard, FinArg-ARC, MicroF1· TheFinAI public numeric results66%
2DeepSeek V3published-demo-leaderboard, FinArg-ARC, MicroF1· TheFinAI public numeric results65.8%
3Qwen 2 (72B)published-demo-leaderboard, FinArg-ARC, MicroF1· TheFinAI public numeric results63%
4Qwen2.5 7Bpublished-demo-leaderboard, FinArg-ARC, MicroF1· TheFinAI public numeric results62.6%
5gemini-1.0-propublished-demo-leaderboard, FinArg-ARC, MicroF1· TheFinAI public numeric results60%
6Mixtral 8X7b V0.1published-demo-leaderboard, FinArg-ARC, MicroF1· TheFinAI public numeric results57%
7code-llama-7b-instructpublished-demo-leaderboard, FinArg-ARC, MicroF1· TheFinAI public numeric results57%
8Llama 3.1 70Bpublished-demo-leaderboard, FinArg-ARC, MicroF1· TheFinAI public numeric results55%
9Llama 3.1 8Bpublished-demo-leaderboard, FinArg-ARC, MicroF1· TheFinAI public numeric results55%
10DeepSeek R1 Distill Qwen 7Bpublished-demo-leaderboard, FinArg-ARC, MicroF1· TheFinAI public numeric results53.1%
11Qwen2 7B Instructpublished-demo-leaderboard, FinArg-ARC, MicroF1· TheFinAI public numeric results53%
12Qwen2.5 7B Instructpublished-demo-leaderboard, FinArg-ARC, MicroF1· TheFinAI public numeric results49.7%
13Llama 3.2 3Bpublished-demo-leaderboard, FinArg-ARC, MicroF1· TheFinAI public numeric results48.9%
14codellama-7bpublished-demo-leaderboard, FinArg-ARC, MicroF1· TheFinAI public numeric results47.3%
15DeepSeek R1 Distill Qwen 1.5Bpublished-demo-leaderboard, FinArg-ARC, MicroF1· TheFinAI public numeric results44.3%
16GPT-4 Turbopublished-demo-leaderboard, FinArg-ARC, MicroF1· TheFinAI public numeric results40%
17GPT-3.5 Turbopublished-demo-leaderboard, FinArg-ARC, MicroF1· TheFinAI public numeric results39%
18Llama 2 70bpublished-demo-leaderboard, FinArg-ARC, MicroF1· TheFinAI public numeric results36%
19Llama-2-7b-chat-hfpublished-demo-leaderboard, FinArg-ARC, MicroF1· TheFinAI public numeric results27%

Data credits: The Fin AI. Published FinBen result summaries and Open Greek Financial LLM Leaderboard aggregates. Results licence not separately stated; fractional FinBen success metrics converted to percentages.