This may take a moment. We're preparing the latest results.
This may take a moment. We're preparing the latest results.
Free credits included. No credit card required.
Phi-4-mini-instruct is a lightweight open model built upon synthetic data and filtered publicly available websites with a focus on high-quality, reasoning dense data. The model belongs to the Phi-4 model family and supports 128K token context length. It employs new architecture for efficiency, larger vocabulary for multilingual support, and better post-training techniques for instruction following and function calling.
Evaluations
6
across 6 benchmarks
Median latency
321ms
Context length
128k
tokens
Cost
$0.08 · $0.35
input · output per 1M tokens
Key takeaways
Technical specifications