Dataset · csoai
Cryptographically signed evaluation cards for open AI models
csoai/gspc-hub-cards
Tabular dataset with cryptographically signed benchmark results of open models across various metrics.
What's inside
Contains between 1,000 and 10,000 tabular records in English where each row corresponds to a specific model, benchmark axis, and evaluation date. The data represents measurement cards from the Council of AI project equipped with an Ed25519 digital signature for independent authenticity verification.
What you can do with it
- Analyzing and comparing public benchmark results of open models
- Building verifiable dashboards and LLM progress monitoring systems
- Tasks
- Other
- Languages
- English
- Size
- 1K–10K rows
- Downloads this month
- 25.4K
- Likes
- 0
How to load it
from datasets import load_dataset
ds = load_dataset("csoai/gspc-hub-cards")