Skip to content

Benchmark

FinBen SC F1

Scores are percentages. Higher is better.

Models measured
19
Scores recorded
19
Sources
1
IQ weight
0.95

Score distribution

74.9%
Mean
82%
Median
16.5%
Std dev
60%
10th percentile
87.3%
90th percentile
19.5%
Min
19.5%89.2%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.95 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
1Qwen2.5 7B Instructpublished-demo-leaderboard, SC, F1· TheFinAI public numeric results89.2%
2DeepSeek V3published-demo-leaderboard, SC, F1· TheFinAI public numeric results87.9%
3Llama 3.2 3Bpublished-demo-leaderboard, SC, F1· TheFinAI public numeric results87.3%
4Llama 3.1 70Bpublished-demo-leaderboard, SC, F1· TheFinAI public numeric results87%
5Llama-2-7b-chat-hfpublished-demo-leaderboard, SC, F1· TheFinAI public numeric results85%
6Mistral (7B) Instruct v0.3published-demo-leaderboard, SC, F1· TheFinAI public numeric results83.8%
7codellama-7bpublished-demo-leaderboard, SC, F1· TheFinAI public numeric results83.3%
8Llama 3.1 8Bpublished-demo-leaderboard, SC, F1· TheFinAI public numeric results83%
9Mixtral 8X7b V0.1published-demo-leaderboard, SC, F1· TheFinAI public numeric results83%
10Qwen 2 (72B)published-demo-leaderboard, SC, F1· TheFinAI public numeric results82%
11GPT-4 Turbopublished-demo-leaderboard, SC, F1· TheFinAI public numeric results81%
12GPT-3.5 Turbopublished-demo-leaderboard, SC, F1· TheFinAI public numeric results80%
13gemini-1.0-propublished-demo-leaderboard, SC, F1· TheFinAI public numeric results74%
14Qwen2.5 7Bpublished-demo-leaderboard, SC, F1· TheFinAI public numeric results73.9%
15DeepSeek R1 Distill Qwen 7Bpublished-demo-leaderboard, SC, F1· TheFinAI public numeric results68.2%
16Llama 2 70bpublished-demo-leaderboard, SC, F1· TheFinAI public numeric results61%
17Qwen2 7B Instructpublished-demo-leaderboard, SC, F1· TheFinAI public numeric results60%
18code-llama-7b-instructpublished-demo-leaderboard, SC, F1· TheFinAI public numeric results53.4%
19DeepSeek R1 Distill Qwen 1.5Bpublished-demo-leaderboard, SC, F1· TheFinAI public numeric results19.5%

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.