Open-weight instruction model for adaptable chat and self-hosted production workloads
Phi-4 by microsoft is offered by 4 providers on this page. Its official list price is $0.125 per 1M input tokens and $0.50 per 1M output tokens. The lowest paid channel is OpenRouter at $0.07 / $0.14 per 1M, about 1.8× below the list price. That channel is a third-party gateway, so confirm its availability and rate limits before depending on it.
Artificial Analysis rates it 5.9 on the Intelligence Index. Against its lowest blended price of $0.087 per 1M, that is roughly 67 index points per dollar, which is the value ratio the leaderboards rank on. Median output speed is 39 tokens per second, with 2.59s to the first token.
The context window is 128,000 tokens at the reference host, but hosts report different limits, from 16,384 to 128,000, so the usable window depends on the provider you pick. It supports structured output. The weights are open, so it can also be self-hosted or served through a gateway of your choice. Providers report the capability flags inconsistently, so verify a specific feature against the host you plan to use. Its training knowledge cuts off at 2023-10.
Quality is independently evaluated by Artificial Analysis. Speed/latency are model-level medians.
| Provider | Tier | Input | Output | Cache read | Cache write | Context | Output limit | Status |
|---|---|---|---|---|---|---|---|---|
| OpenRouter | Gateway | $0.07 | $0.14 | — | — | 16,384 ⚠ | 14,745 | |
| Kilo Gateway | Gateway | $0.07 | $0.14 | — | — | 16,384 ⚠ | 14,745 | |
| AzureOfficial | Cloud | $0.13 | $0.50 | — | — | 128,000 | 4,096 | |
| Azure Cognitive ServicesOfficial | Cloud | $0.13 | $0.50 | — | — | 128,000 | 4,096 |
Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.
Individual evaluations run by Artificial Analysis, on the same reasoning tier as the intelligence score above. Each benchmark has its own task set and harness, so rows are not comparable with one another. The Intelligence Index above draws on a different, newer set of evaluations.