Phi 4
Overview
phi-4 is a state-of-the-art open model built to excel at advanced reasoning, coding, and knowledge tasks. It leverages a blend of synthetic data, filtered web data, academic texts, and supervised fine-tuning for precision, alignment, and safety.
Phi 4 was released on December 12, 2024. API access is available through DeepInfra.
Performance
Timeline
ReleasedUnknown
Knowledge CutoffUnknown
Specifications
Parameters
14.7B
License
MIT
Training Data
Unknown
Benchmarks
Phi 4 Performance Across Datasets
Scores sourced from the model's scorecard, paper, or official blog posts
llm-stats.com - Mon Dec 22 2025Notice missing or incorrect data?Start an Issue discussion→
Pricing
Pricing, performance, and capabilities for Phi 4 across different providers:
| Provider | Input ($/M) | Output ($/M) | Max Input | Max Output | Latency (s) | Throughput | Quantization | Input | Output |
|---|---|---|---|---|---|---|---|---|---|
 DeepInfra | $0.07 | $0.14 | 16.0K | 16.0K | 0.2 | 33.0 tok/s | — | Text Image Audio Video | Text Image Audio Video |
API Access
API Access Coming Soon
API access for Phi 4 will be available soon through our gateway.
Recent Posts
Recent Reviews
FAQ
Common questions about Phi 4
Phi 4 was released on December 12, 2024.
Phi 4 has 14.7 billion parameters.