Skip to content

Benchmark

FinBen FPB Acc

Scores are percentages. Higher is better.

Models measured
16
Scores recorded
16
Sources
1
IQ weight
0.54

Score distribution

57.7%
Mean
66.2%
Median
18.4%
Std dev
38.9%
10th percentile
77%
90th percentile
14.6%
Min
14.6%78%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.54 for chat models, measured over 16 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1GPT-3.5 Turbopublished-demo-leaderboard, FPB, Acc· TheFinAI public numeric results78%
2gemini-1.0-propublished-demo-leaderboard, FPB, Acc· TheFinAI public numeric results77%
3Qwen2.5 7Bpublished-demo-leaderboard, FPB, Acc· TheFinAI public numeric results76.1%
4GPT-4 Turbopublished-demo-leaderboard, FPB, Acc· TheFinAI public numeric results76%
5Llama 2 70bpublished-demo-leaderboard, FPB, Acc· TheFinAI public numeric results72%
6codellama-7bpublished-demo-leaderboard, FPB, Acc· TheFinAI public numeric results71.9%
7Llama 3.2 3Bpublished-demo-leaderboard, FPB, Acc· TheFinAI public numeric results69.2%
8Mistral (7B) Instruct v0.3published-demo-leaderboard, FPB, Acc· TheFinAI public numeric results66.2%
9code-llama-7b-instructpublished-demo-leaderboard, FPB, Acc· TheFinAI public numeric results63.4%
10Qwen2 7B Instructpublished-demo-leaderboard, FPB, Acc· TheFinAI public numeric results52%
11Qwen2.5 7B Instructpublished-demo-leaderboard, FPB, Acc· TheFinAI public numeric results46.9%
12DeepSeek R1 Distill Qwen 7Bpublished-demo-leaderboard, FPB, Acc· TheFinAI public numeric results42.7%
13Llama-2-7b-chat-hfpublished-demo-leaderboard, FPB, Acc· TheFinAI public numeric results41%
14DeepSeek V3published-demo-leaderboard, FPB, Acc· TheFinAI public numeric results38.9%
15Mixtral 8X7b V0.1published-demo-leaderboard, FPB, Acc· TheFinAI public numeric results37%
16DeepSeek R1 Distill Qwen 1.5Bpublished-demo-leaderboard, FPB, Acc· TheFinAI public numeric results14.6%

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.