Microsoft
Phi 4
Released 2025-01-10 · Knowledge up to 2024-06-30
Microsoft Research 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...
You send / 1M
$0.070
tokens in
It replies / 1M
$0.140
tokens out
Memory
16K
tokens it can read at once
Typical request
$0.0002
1,000 words in, 400 out
Data privacy
No training or retention: Yes. At least one provider serves it with zero data retention.
| Provider | You send / 1M | It replies / 1M | Uptime | Privacy policy |
|---|---|---|---|---|
| DeepInfra · bf16 | $0.056 | $0.112 | 100% | Read |
| DeepInfra · bf16 | $0.070 | $0.140 | 100% | Read |
Based on OpenRouter and Vercel AI Gateway data. Confirm terms with the provider before sending sensitive data.
What it can do
- Reads: text
- Writes: text
- Reasoning: No
- Tool calling: No
- Longest reply: 15K tokens
- Intelligence score: 5.9 / 100 (Artificial Analysis)
Similar models
- Nova Micro 1.0Amazon · $0.035 in / $0.140 out per 1M
- Schematron V2 TurboInference Net · $0.030 in / $0.150 out per 1M
- Command R7B (12-2024)Cohere · $0.037 in / $0.150 out per 1M
- Hy-MT2-1.8BTencent · $0.044 in / $0.177 out per 1M
- MythoMax 13BGryphe · $0.080 in / $0.110 out per 1M
- Qwen2.5 7B InstructQwen · $0.100 in / $0.200 out per 1M