Price updated Mar 18, 2026: input price per 1m changed from $0.0600 to $0.0650/1M tokens (+8.3%)
Phi 4 API Cost Calculator
Estimate your monthly API costs for Phi 4 from Microsoft.
Pricing per 1M tokens
Input
$0.0650
Output
$0.1400
16.4K context16.4K max outputStreaming
Pricing last verified: Mar 21, 2026
30-Day Price History
Solid line: input price · Dashed line: output price
Estimate Your Costs
What are you building?
Cost Breakdown
Input cost/request$0.000033
Output cost/request$0.000042
Cost per request$0.000074
Daily cost (1,000 requests)$0.0745
Monthly estimate$2.24
Annual estimate$26.82
Compare with Similar Models
| Model | Input/1M | Output/1M |
|---|---|---|
| WizardLM-2 8x22B Microsoft | $0.6200 | $0.6200 |
| Ministral 3 3B 2512 Mistralai | $0.1000 | $0.1000 |
| Nemotron Nano 9B V2 Nvidia | $0.0400 | $0.1600 |
| Qwen3.5-9B Qwen | $0.0500 | $0.1500 |
| GLM 4 32B Z Ai | $0.1000 | $0.1000 |
Frequently Asked Questions
How much does Phi 4 cost per token?
Phi 4 is priced at $0.0650 per 1M input tokens and $0.1400 per 1M output tokens. Use the calculator above to estimate your monthly costs based on your usage volume.
What is the context window for Phi 4?
Phi 4 has a context window of 16.4K tokens (16,384 tokens). The maximum output length is 16.4K tokens.
How does Phi 4 compare to similar models?
Phi 4 costs $0.0650 input and $0.1400 output per 1M tokens. GLM 4 32B is less expensive at $0.1000 input and $0.1000 output per 1M tokens. See the "Compare with Similar Models" table above, or visit the GLM 4 32B page at /tools/cost-calculator/z-ai/glm-4-32b.