Skip to content

Benchmark

FinBen EFPA ACC

Scores are percentages. Higher is better.

Models measured
13
Scores recorded
13
Sources
1
IQ weight
1.30

Score distribution

28%
Mean
25.9%
Median
4.7%
Std dev
25.4%
10th percentile
34%
90th percentile
25%
Min
25%41.7%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.30 for chat models, measured over 13 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Qwen2.5 7Bpublished-demo-leaderboard, EFPA, Acc· TheFinAI public numeric results41.7%
2GPT-4 Turbopublished-demo-leaderboard, EFPA, ACC· TheFinAI public numeric results34%
3GPT-3.5 Turbopublished-demo-leaderboard, EFPA, ACC· TheFinAI public numeric results31%
4Qwen2 7B Instructpublished-demo-leaderboard, EFPA, ACC· TheFinAI public numeric results27%
5code-llama-7b-instructpublished-demo-leaderboard, EFPA, Acc· TheFinAI public numeric results26.3%
6Llama-2-7b-chat-hfpublished-demo-leaderboard, EFPA, ACC· TheFinAI public numeric results26%
7Mistral (7B) Instruct v0.3published-demo-leaderboard, EFPA, Acc· TheFinAI public numeric results25.9%
8codellama-7bpublished-demo-leaderboard, EFPA, Acc· TheFinAI public numeric results25.4%
9DeepSeek R1 Distill Qwen 7Bpublished-demo-leaderboard, EFPA, ACC· TheFinAI public numeric results25.4%
10DeepSeek V3published-demo-leaderboard, EFPA, ACC· TheFinAI public numeric results25.4%
11Llama 3.2 3Bpublished-demo-leaderboard, EFPA, Acc· TheFinAI public numeric results25.4%
12Qwen2.5 7B Instructpublished-demo-leaderboard, EFPA, Acc· TheFinAI public numeric results25.4%
13Mixtral 8X7b V0.1published-demo-leaderboard, EFPA, ACC· TheFinAI public numeric results25%

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.