Microsoft: Phi 4 vs OpenAI: GPT-5 Mini
Microsoft: Phi 4 is cheaper than OpenAI: GPT-5 Mini by about 87% on a blended (75% input / 25% output) basis. Microsoft: Phi 4 charges $0.070000 / $0.140000 per 1M input/output tokens; OpenAI: GPT-5 Mini charges $0.250000 / $2.000000.
| Field | Microsoft: Phi 4 | OpenAI: GPT-5 Mini |
|---|---|---|
| Input $/Mtok | $0.070000 | $0.250000 |
| Output $/Mtok | $0.140000 | $2.000000 |
| Cache read $/Mtok | – | $0.025000 |
| Cache write $/Mtok | – | – |
| Input $/Mchar | n/a | 0.070428 |
| Output $/Mchar | n/a | 0.563428 |
| Context window | 16384 | 400000 |
| 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: GPT-5 Mini |
|---|---|---|
| Coding agent | $0.74 | $3.38 |
| RAG chat | $0.81 | $5.13 |
| Summarisation (batch) | $0.77 | $4.25 |
| Classification | $0.71 | $2.85 |
| Chat assistant | $0.98 | $9.50 |
| Content generation | $1.26 | $16.50 |
Cite this page
Microsoft: Phi 4 vs OpenAI: GPT-5 Mini: $0.070000/$0.140000 vs $0.250000/$2.000000 per 1M tokens as of 2026-10-04. Token Barometer by Dejan Strbad, https://tokenbarometer.com, data CC BY 4.0. https://tokenbarometer.com/compare/microsoft_phi-4-vs-openai_gpt-5-mini
Microsoft: Phi 4 page | OpenAI: GPT-5 Mini page | Open in calculator