phi-3-small 7.4B
#193 of 274 scored.
MicrosoftApr 23, 2024Open weights
Capabilities Index
Range 114.8–124.4
121.8
Context
Tokens
—
Input
Per 1M tokens
—
Output
Per 1M tokens
—
Benchmarks
No independent scores yet.
Epoch AI hasn't run phi-3-small 7.4B yet. Its scores show up here the morning after they do.
Everything else Epoch has for it
- ARC AI287.6%
- OpenBookQA84.0%
- BBH72.1%
- HellaSwag69.3%
- MMLU67.6%
- Winogrande63.0%
- TriviaQA58.1%
- ANLI37.1%
On the index
Around it.
Sources