Model Library
MicrosoftReleased February 26, 2024Phi-4 Mini
Microsoft · Released February 26, 2024
Intelligence score
6
At a glance
- Intelligence
- 6
- Speed
- 43.7tokens/sec
- Latency
- 840mstime to first token
- Input price
- Not published
- Output price
- Not published
- Context window
- 130Ktokens
- Parameters
- 3.84billions
Summary
Comparison summary
Phi-4 Mini Instruct scores 6 on the Artificial Analysis Intelligence Index, placing it well above average among other open weight non-reasoning models of similar size (median: 3). Phi-4 Mini Instruct generates output at 43.7 tokens per second (based on Microsoft's API), which is at the lower end compared to other open weight non-reasoning models of similar size (median: 85.1 t/s).
Specifications
Technical specifications
| Model name | Phi-4 Mini |
|---|---|
| Provider | Microsoft |
| Release date | February 26, 2024 |
| Intelligence score | 6 |
| Speed | 43.7 tokens/sec |
| Latency (TTFT) | 0.84s |
| Input price | Not published |
| Output price | Not published |
| Context window | 130,000 tokens |
| Parameters | 3.84B |
| Reasoning | No |
| Input modalities | text |
| Output modalities | — |
| Open weights | Yes |
| License | MIT |
| API availability | — |
| Source | Model Intelligence legacy profile: phi-4-mini |
FAQ · 18
Frequently asked questions
Phi-4 Mini Instruct was released on February 26, 2024.
Phi-4 Mini Instruct was created by Microsoft.
Phi-4 Mini Instruct scores 6 on the Artificial Analysis Intelligence Index, placing it well above average among other open weight non-reasoning models of similar size (median: 3).
Phi-4 Mini Instruct generates output at 43.7 tokens per second (based on Microsoft's API), which is at the lower end compared to other open weight non-reasoning models of similar size (median: 85.1 t/s).
Phi-4 Mini Instruct has a time to first token (TTFT) of 0.84s (based on Microsoft's API), which is better than average compared to other open weight non-reasoning models of similar size (median: 0.84s).
When evaluated on the Intelligence Index, Phi-4 Mini Instruct generated 32M output tokens, which is somewhat higher than average compared to other open weight non-reasoning models of similar size (median: 27M).
No, Phi-4 Mini Instruct is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.
Phi-4 Mini Instruct supports text only input.
Phi-4 Mini Instruct supports text only output.
No, Phi-4 Mini Instruct does not support image input. It can only process text.
No, Phi-4 Mini Instruct is not multimodal. It only supports text only input.
Phi-4 Mini Instruct has a context window of 130k tokens. This determines how much text and conversation history the model can process in a single request.
Yes, Phi-4 Mini Instruct is open weights. The model weights are publicly available and can be downloaded for self-hosting.
Phi-4 Mini Instruct has 3.84 billion parameters.
Phi-4 Mini Instruct is released under the MIT license. This license allows commercial use.
Phi-4 Mini Instruct achieves a score of 6 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.
Yes, Phi-4 Mini Instruct is available via API through 2 providers.
Phi-4 Mini Instruct is available through 2 API providers.