Phi 4 pricing & benchmarks
Released Jan 10, 2025 · served by 1 provider
What Phi 4 is
[Microsoft Research](/microsoft) Phi-4 is designed to perform well in complex reasoning tasks and can operate efficiently in situations with limited memory or where quick responses are needed. At 14 billion...
- Model ID:
microsoft/phi-4 - Modality: text->text
- Knowledge cutoff: 2024-06-30
- Tool calling: not supported · Structured output: supported
- Weights:
microsoft/phi-4on Hugging Face
Providers and prices for Phi 4
The same model costs different amounts depending on who serves it. Prices are per 1M tokens, cheapest first. Uptime and throughput are OpenRouter's rolling measurements, not vendor claims.
| Provider | Input | Output | Context | Quant | Uptime 24h | Throughput |
|---|---|---|---|---|---|---|
| DeepInfracheapest | $0.070 | $0.140 | 16K | bf16 | 100.0% | — |
FAQ
How much does the Phi 4 API cost?
$0.070 per 1M input tokens and $0.140 per 1M output tokens. A typical workload of 1M input + 300K output tokens costs about $0.11.
Is Phi 4 free?
No. It is a paid model starting at $0.070 per 1M input tokens, though some providers offer trial credits.
Which provider is cheapest for Phi 4?
DeepInfra at $0.070 input / $0.140 output per 1M tokens (bf16 quantization).
Can I self-host Phi 4?
Yes — weights are published as microsoft/phi-4 on Hugging Face, so you can run it on your own hardware.