Skip to content

Benchmark

FinBen Headlines AvgF1

Scores are percentages. Higher is better.

Models measured
19
Scores recorded
19
Sources
1
IQ weight
0.22

Score distribution

65.2%
Mean
60%
Median
8%
Std dev
60%
10th percentile
77%
90th percentile
60%
Min
60%86%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.22 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
1GPT-4 Turbopublished-demo-leaderboard, Headlines, AvgF1· TheFinAI public numeric results86%
2gemini-1.0-propublished-demo-leaderboard, Headlines, AvgF1· TheFinAI public numeric results78%
3GPT-3.5 Turbopublished-demo-leaderboard, Headlines, AvgF1· TheFinAI public numeric results77%
4Qwen2.5 7Bpublished-demo-leaderboard, Headlines, AvgF1· TheFinAI public numeric results74.9%
5Llama-2-7b-chat-hfpublished-demo-leaderboard, Headlines, AvgF1· TheFinAI public numeric results72%
6Mistral (7B) Instruct v0.3published-demo-leaderboard, Headlines, AvgF1· TheFinAI public numeric results68.5%
7Llama 2 70bpublished-demo-leaderboard, Headlines, AvgF1· TheFinAI public numeric results63%
8Llama 3.2 3Bpublished-demo-leaderboard, Headlines, AvgF1· TheFinAI public numeric results60.3%
9Llama 3.1 70Bpublished-demo-leaderboard, Headlines, AvgF1· TheFinAI public numeric results60%
10Llama 3.1 8Bpublished-demo-leaderboard, Headlines, AvgF1· TheFinAI public numeric results60%
11Mixtral 8X7b V0.1published-demo-leaderboard, Headlines, AvgF1· TheFinAI public numeric results60%
12Qwen 2 (72B)published-demo-leaderboard, Headlines, AvgF1· TheFinAI public numeric results60%
13Qwen2 7B Instructpublished-demo-leaderboard, Headlines, AvgF1· TheFinAI public numeric results60%
14code-llama-7b-instructpublished-demo-leaderboard, Headlines, AvgF1· TheFinAI public numeric results60%
15codellama-7bpublished-demo-leaderboard, Headlines, AvgF1· TheFinAI public numeric results60%
16DeepSeek R1 Distill Qwen 1.5Bpublished-demo-leaderboard, Headlines, AvgF1· TheFinAI public numeric results60%
17DeepSeek R1 Distill Qwen 7Bpublished-demo-leaderboard, Headlines, AvgF1· TheFinAI public numeric results60%
18DeepSeek V3published-demo-leaderboard, Headlines, AvgF1· TheFinAI public numeric results60%
19Qwen2.5 7B Instructpublished-demo-leaderboard, Headlines, AvgF1· TheFinAI public numeric results60%

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.