Microsoft model product
Phi 4 Mini
8B parameters) open model built upon synthetic data and filtered web data, focusing on high-quality reasoning.
Updated Aug 17, 2026. Default version: Phi 4 Mini
LLMBoard Score0.0Phi 4 Mini
Coverage60%17 benchmark families
Context window128KTokens
Official input price$0.075Azure API
Phi 4 Mini Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Phi 4 Mini Benchmark Results
Benchmark scores for Phi 4 Mini.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkOpenBookQA | Score79.2% | Rank03 | Participants5 | Percentile50.0% | EvidenceC | Evaluated |
| BenchmarkSocial IQa | Score72.5% | Rank03 | Participants9 | Percentile75.0% | EvidenceC | Evaluated |
| BenchmarkTruthfulQA | Score66.4% | Rank04 | Participants18 | Percentile82.3% | EvidenceC | Evaluated |
| BenchmarkMultilingual MMLU | Score49.3% | Rank05 | Participants5 | Percentile0.0% | EvidenceC | Evaluated |
| BenchmarkBoolQ | Score81.2% | Rank07 | Participants10 | Percentile33.3% | EvidenceC | Evaluated |
| BenchmarkPIQA | Score77.6% | Rank10 | Participants11 | Percentile10.0% | EvidenceC | Evaluated |
| BenchmarkBIG-Bench Hard | Score70.4% | Rank12 | Participants21 | Percentile45.0% | EvidenceC | Evaluated |
| BenchmarkARC-C | Score83.7% | Rank15 | Participants34 | Percentile57.6% | EvidenceC | Evaluated |
| BenchmarkWinogrande | Score67.0% | Rank19 | Participants22 | Percentile14.3% | EvidenceC | Evaluated |
| BenchmarkArena Hard | Score32.8% | Rank24 | Participants26 | Percentile8.0% | EvidenceC | Evaluated |
| BenchmarkMGSM | Score63.9% | Rank24 | Participants31 | Percentile23.3% | EvidenceC | Evaluated |
| BenchmarkHellaSwag | Score69.1% | Rank26 | Participants27 | Percentile3.9% | EvidenceC | Evaluated |
| BenchmarkGSM8k | Score88.6% | Rank32 | Participants48 | Percentile34.0% | EvidenceC | Evaluated |
| BenchmarkMATH | Score64.0% | Rank45 | Participants71 | Percentile37.1% | EvidenceC | Evaluated |
| BenchmarkMMLU | Score67.3% | Rank90 | Participants101 | Percentile11.0% | EvidenceC | Evaluated |
| BenchmarkMMLU-Pro | Score52.8% | Rank115 | Participants134 | Percentile14.3% | EvidenceC | Evaluated |
| BenchmarkGPQA | Score25.2% | Rank233 | Participants239 | Percentile2.5% | EvidenceC | Evaluated |
Phi 4 Mini Arena Results
Preference and agent-evaluation results for the default version.
No Arena results
The default version does not have a matching Arena result yet.
Phi 4 Mini Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $0.075 input, $0.30 output per 1M
- Official provider
- Azure
- Lowest third-party
- From $0.075 input, $0.30 output per 1M via Azure Cognitive Services
- Tracked offerings
- 4
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderNvidia | Provider model IDmicrosoft/phi-4-mini-instruct | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context131.1K | Updated |
| ProviderAzure Cognitive Services | Provider model IDphi-4-mini | Regionglobal | Input / 1M$0.075 | Output / 1M$0.30 | Context128K | Updated |
| ProviderAzure | Provider model IDphi-4-mini | Regionglobal | Input / 1M$0.075 | Output / 1M$0.30 | Context128K | Updated |
| ProviderNanoGPT | Provider model IDphi-4-mini-instruct | Regionglobal | Input / 1M$0.17 | Output / 1M$0.68 | Context128K | Updated |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Phi 4 Mini Runtime Performance
Provider-specific output speed and catalog latency for Phi 4 Mini. Runtime does not affect the capability score.
No runtime data
Browse runtime rankingsNo provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Phi 4 Mini Specifications
Technical details for the model's default version.
- Version
- Phi 4 Mini
- Released
- Feb 1, 2025
- Knowledge cutoff
- Jun 1, 2024
- Parameters
- 3.8B
- Context window
- 128K
- Max output
- 4.1K
- Inputs
- text
- Outputs
- text
- Open weights
- Yes
- License
- MIT
Phi 4 Mini Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionPhi 4 Mini | Released | LLMBoard0.0 | Parameters3.8B | Context128K | Max output4.1K | Open weightsYes | LicenseMIT |
Phi 4 Mini vs nearby models
Open a comparison with the three ranked models immediately above and below this model.
Models similar to Phi 4 Mini
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
MI
Phi 3.5 vision
Microsoft
MI
Phi 3.5 mini
Microsoft
MI
Phi 3.5 MoE
Microsoft
MI
Phi 4 multimodal
Microsoft
MI
Phi 4
Microsoft
MI
Phi 4 Mini Reasoning
Microsoft
What is Phi 4 Mini?
Key information about Phi 4 Mini and its available data.
8B parameters) open model built upon synthetic data and filtered web data, focusing on high-quality reasoning. It supports a 128K token context length and is enhanced for instruction adherence and safety via supervised fine-tuning and direct preference optimization.
Data as of 2026-08-17.
FAQ
Common questions about Phi 4 Mini.
When was Phi 4 Mini released?
Phi 4 Mini's default version was released on Feb 1, 2025.
How much does Phi 4 Mini cost?
Phi 4 Mini's official API price is $0.075 per million input tokens and $0.30 per million output tokens via Azure. The lowest tracked third-party offer starts at $0.075 input and $0.30 output via Azure Cognitive Services.
Who created Phi 4 Mini?
Phi 4 Mini was created by Microsoft.
What is the context window for Phi 4 Mini?
The default version has a 128K token context window.
Is Phi 4 Mini open weight?
Yes. The default version is marked as open weight under MIT.
How many API providers offer Phi 4 Mini?
4 provider offerings are linked to the default version.
What models should I compare Phi 4 Mini with?
Nearby ranked alternatives include Llama 3.1 8B, Qwen2 7B, ERNIE 4.5.