Skip to content

Benchmark

FinBen MultiFin MicroF1

Scores are percentages. Higher is better.

Models measured
18
Scores recorded
18
Sources
1
IQ weight
2.06

Score distribution

49.8%
Mean
59.9%
Median
18.2%
Std dev
20%
10th percentile
65%
90th percentile
12%
Min
12%69%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.06 for chat models, measured over 18 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Llama 3.1 70Bpublished-demo-leaderboard, MultiFin, MicroF1· TheFinAI public numeric results69%
2DeepSeek V3published-demo-leaderboard, MultiFin, MicroF1· TheFinAI public numeric results66.7%
3GPT-4 Turbopublished-demo-leaderboard, MultiFin, MicroF1· TheFinAI public numeric results65%
4Llama 2 70bpublished-demo-leaderboard, MultiFin, MicroF1· TheFinAI public numeric results63%
5gemini-1.0-propublished-demo-leaderboard, MultiFin, MicroF1· TheFinAI public numeric results62%
6Llama 3.1 8Bpublished-demo-leaderboard, MultiFin, MicroF1· TheFinAI public numeric results62%
7Qwen2.5 7B Instructpublished-demo-leaderboard, MultiFin, MicroF1· TheFinAI public numeric results61.9%
8Qwen2.5 7Bpublished-demo-leaderboard, MultiFin, MicroF1· TheFinAI public numeric results60.7%
9codellama-7bpublished-demo-leaderboard, MultiFin, MicroF1· TheFinAI public numeric results59.9%
10GPT-3.5 Turbopublished-demo-leaderboard, MultiFin, MicroF1· TheFinAI public numeric results59%
11Qwen 2 (72B)published-demo-leaderboard, MultiFin, MicroF1· TheFinAI public numeric results55%
12Mistral (7B) Instruct v0.3published-demo-leaderboard, MultiFin, MicroF1· TheFinAI public numeric results54.9%
13Qwen2 7B Instructpublished-demo-leaderboard, MultiFin, MicroF1· TheFinAI public numeric results39%
14Mixtral 8X7b V0.1published-demo-leaderboard, MultiFin, MicroF1· TheFinAI public numeric results37%
15code-llama-7b-instructpublished-demo-leaderboard, MultiFin, MicroF1· TheFinAI public numeric results36.2%
16Llama-2-7b-chat-hfpublished-demo-leaderboard, MultiFin, MicroF1· TheFinAI public numeric results20%
17DeepSeek R1 Distill Qwen 1.5Bpublished-demo-leaderboard, MultiFin, MicroF1· TheFinAI public numeric results13.5%
18DeepSeek R1 Distill Qwen 7Bpublished-demo-leaderboard, MultiFin, MicroF1· TheFinAI public numeric results12%

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.