Microsoft: Phi 4 vs OpenAI: o4 Mini
Microsoft: Phi 4 is cheaper than OpenAI: o4 Mini by about 95% on a blended (75% input / 25% output) basis. Microsoft: Phi 4 charges $0.070000 / $0.140000 per 1M input/output tokens; OpenAI: o4 Mini charges $1.100000 / $4.400000.
| Field | Microsoft: Phi 4 | OpenAI: o4 Mini |
|---|---|---|
| Input $/Mtok | $0.070000 | $1.100000 |
| Output $/Mtok | $0.140000 | $4.400000 |
| Cache read $/Mtok | – | $0.275000 |
| Cache write $/Mtok | – | – |
| Input $/Mchar | n/a | 0.309885 |
| Output $/Mchar | n/a | 1.239541 |
| Context window | 16384 | 200000 |
| Tokenizer | – | o200k |
Price history
Same workload, monthly cost
Illustrative monthly cost at 10M tokens/month, split per workload preset. Not a substitute for the calculator, which uses your own volume, cache and batch assumptions.
| Workload | Microsoft: Phi 4 | OpenAI: o4 Mini |
|---|---|---|
| Coding agent | $0.74 | $12.65 |
| RAG chat | $0.81 | $15.95 |
| Summarisation (batch) | $0.77 | $14.30 |
| Classification | $0.71 | $11.66 |
| Chat assistant | $0.98 | $24.20 |
| Content generation | $1.26 | $37.40 |
Cite this page
Microsoft: Phi 4 vs OpenAI: o4 Mini: $0.070000/$0.140000 vs $1.100000/$4.400000 per 1M tokens, as of 2026-10-04 (Token Barometer, https://tokenbarometer.com/compare/microsoft_phi-4-vs-openai_o4-mini).
Microsoft: Phi 4 page | OpenAI: o4 Mini page | Open in calculator