Skip to content

Benchmark

FinBen NER EntityF1

Scores are percentages. Higher is better.

Models measured
19
Scores recorded
19
Sources
1
IQ weight
0.56

Score distribution

21.6%
Mean
14%
Median
24%
Std dev
2.9%
10th percentile
61%
90th percentile
1.8%
Min
1.8%83%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.56 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, NER, EntityF1· TheFinAI public numeric results83%
2GPT-3.5 Turbopublished-demo-leaderboard, NER, EntityF1· TheFinAI public numeric results77%
3gemini-1.0-propublished-demo-leaderboard, NER, EntityF1· TheFinAI public numeric results61%
4Qwen2.5 7B Instructpublished-demo-leaderboard, NER, EntityF1· TheFinAI public numeric results25.6%
5Qwen2.5 7Bpublished-demo-leaderboard, NER, EntityF1· TheFinAI public numeric results24.5%
6Mixtral 8X7b V0.1published-demo-leaderboard, NER, EntityF1· TheFinAI public numeric results24%
7Llama-2-7b-chat-hfpublished-demo-leaderboard, NER, EntityF1· TheFinAI public numeric results18%
8code-llama-7b-instructpublished-demo-leaderboard, NER, EntityF1· TheFinAI public numeric results17.4%
9Mistral (7B) Instruct v0.3published-demo-leaderboard, NER, EntityF1· TheFinAI public numeric results14.8%
10Llama 3.1 8Bpublished-demo-leaderboard, NER, EntityF1· TheFinAI public numeric results14%
11DeepSeek R1 Distill Qwen 1.5Bpublished-demo-leaderboard, NER, EntityF1· TheFinAI public numeric results12.1%
12DeepSeek V3published-demo-leaderboard, NER, EntityF1· TheFinAI public numeric results9.6%
13Qwen2 7B Instructpublished-demo-leaderboard, NER, EntityF1· TheFinAI public numeric results7%
14codellama-7bpublished-demo-leaderboard, NER, EntityF1· TheFinAI public numeric results6.1%
15Llama 3.1 70Bpublished-demo-leaderboard, NER, EntityF1· TheFinAI public numeric results5%
16Llama 2 70bpublished-demo-leaderboard, NER, EntityF1· TheFinAI public numeric results4%
17DeepSeek R1 Distill Qwen 7Bpublished-demo-leaderboard, NER, EntityF1· TheFinAI public numeric results2.9%
18Qwen 2 (72B)published-demo-leaderboard, NER, EntityF1· TheFinAI public numeric results2%
19Llama 3.2 3Bpublished-demo-leaderboard, NER, EntityF1· TheFinAI public numeric results1.8%

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.