Skip to content

Benchmark

FinBen ccf MCC

Higher scores are better.

Models measured
14
Scores recorded
14
Sources
1
IQ weight
0.10

Score distribution

-0.001
Mean
0
Median
0.04
Std dev
-0.021
10th percentile
0.021
90th percentile
-0.1
Min
-0.10.102
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.10 for chat models, measured over 14 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, ccf, MCC· TheFinAI public numeric results0.102
2DeepSeek V3published-demo-leaderboard, ccf, MCC· TheFinAI public numeric results0.0214
3Qwen2 7B Instructpublished-demo-leaderboard, ccf, MCC· TheFinAI public numeric results0.01
4Mistral (7B) Instruct v0.3published-demo-leaderboard, ccf, MCC· TheFinAI public numeric results0.0082
5codellama-7bpublished-demo-leaderboard, ccf, MCC· TheFinAI public numeric results0
6Llama 2 70bpublished-demo-leaderboard, ccf, MCC· TheFinAI public numeric results0
7Llama 3.2 3Bpublished-demo-leaderboard, ccf, MCC· TheFinAI public numeric results0
8Llama-2-7b-chat-hfpublished-demo-leaderboard, ccf, MCC· TheFinAI public numeric results0
9Mixtral 8X7b V0.1published-demo-leaderboard, ccf, MCC· TheFinAI public numeric results0
10Qwen2.5 7Bpublished-demo-leaderboard, ccf, MCC· TheFinAI public numeric results0
11gemini-1.0-propublished-demo-leaderboard, ccf, MCC· TheFinAI public numeric results-0.01
12GPT-4 Turbopublished-demo-leaderboard, ccf, MCC· TheFinAI public numeric results-0.02
13DeepSeek R1 Distill Qwen 7Bpublished-demo-leaderboard, ccf, MCC· TheFinAI public numeric results-0.0206
14GPT-3.5 Turbopublished-demo-leaderboard, ccf, MCC· TheFinAI public numeric results-0.1

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.