Phi-4
AI model by Microsoft. Real-time pricing and benchmark data.
Benchmarks
Editorial Analysis
Profile. Phi-4 is a budget-tier option from Microsoft at $0.50/M output tokens. Despite the low price, it scores competitively on measurable benchmarks — making it a candidate for high-volume production workloads where unit cost dominates.
Latency. On latency, Phi-4 reports 39 tok/s output speed and 2590ms time-to-first-token. Output speed is on the slower end; for real-time chat consider pairing with a streaming front-end or routing latency-sensitive requests to a faster model.
Cost model. For a workload of 1M input tokens + 100K output tokens, Phi-4 costs $0.18 per million input-equivalent requests. Multiply by your monthly request volume to project the bill, and compare against the speed/correctness profile above to decide if the trade-off is worth it for your workload.
Editorial summary generated from public pricing and benchmark data (Artificial Analysis API). All numbers are sourced — see the Artificial Analysis methodology for benchmark definitions. Last refreshed 2026-08-24.
Compare with similar models
| Model | Input | Output | Speed |
|---|---|---|---|
Phi-4Current | $0.13 | $0.50 | 39 tok/s |
| Grok 4 Fast (Non-reasoning) | $0.20 | $0.50 | — |
| Grok 4 Fast (Reasoning) | $0.20 | $0.50 | — |
| Grok 3 mini Reasoning (high) | $0.30 | $0.50 | 86 tok/s |
| Qwen2.5 Instruct 72B | $0.47 | $0.49 | — |
| Hy3 | $0.14 | $0.55 | 67 tok/s |