Microsoft › Phi 3 Mini 4K Instruct › Pricing
Phi 3 Mini 4K Instruct pricing
Cheapest host: Featherless · checked Oct 23, 2026
$0.40 / $0.80per 1M tokens in / out
Price sheetper 1M tokens
| Input | $0.40 | Featherless |
| Output | $0.80 | |
| Context | 4K | tokens |
Exact featherless deployment microsoft/Phi-3-mini-4k-instruct; provider contract, not a family-wide guarantee. USD per million input tokens.
In Microsoft's line-upper 1M in / out
| Microsoft Decision 1 | $0.042 / — | −100% |
| Phi 4 | $0.056 / $0.11 | −86% |
| Phi-4 Mini | $0.075 / $0.30 | −74% |
| Phi 4 Reasoning Plus | $0.07 / $0.35 | −72% |
| Phi 3 Medium 4K Instruct | $0.14 / $0.14 | −72% |
| WizardLM 2 8x22B | $0.48 / $0.48 | −4% |
| Phi 3 Mini 4K Instruct | $0.40 / $0.80 | this model |
Cheaper for the same workvs Phi 3 Mini 4K Instruct
| GLM 5.3 Flash unstatedOpen | -53% | score wins 89% | $0.15 / $0.50 |
| Hy3 unstatedOpen | -54% | score wins 88% | $0.13 / $0.53 |
| Gemma 4 31B unstatedOpen | -68% | score wins 88% | $0.10 / $0.34 |
| GPT-6 Luna max | -60% | score wins 87% | $0.068 |
| GPT-5.6 Luna xhigh | -10% | score wins 87% | $0.085 |
Pricing questions
How much does Phi 3 Mini 4K Instruct cost?›
$0.40 per 1M input tokens and $0.80 per 1M output tokens.
Is Phi 3 Mini 4K Instruct free to use?›
There is no free API tier, but the weights are open: you can run it on your own hardware for the cost of compute.
What is a cheaper alternative to Phi 3 Mini 4K Instruct?›
GLM 5.3 Flash at unstated effort costs -53% per task with a score wins 89% on Arena Text · overall.