Microsoft

Phi-3-mini instruct (128k)

Type
language model
Context
128K tokens
Max output
4K
Released
23 April 2024
API string
microsoft/phi-3-mini-128k-instruct
Weights
open

Prices by provider

Provider Input, $ per 1M Output, $ per 1M In our data since
Fireworks AI $0.1 $0.1 31 Jul 2026
Microsoft Foundry $0.13 $0.52 31 Jul 2026

Only the base tier and only per-token prices are shown. Batch, discounted and cached rates, as well as prices per image or per second of video, do not go into this table: they cannot stand in the same column as a price per million tokens.

The date in the last column is the day this price first entered our collection. The price may well be older: before that day we simply were not recording it. It has not changed since — otherwise a new row with a new date would stand in its place.

Measurement results

Task set Result Run conditions Measured by
Arena Score, mathematics 1,089.06 1,076–1,102 — —
Arena Score in English 1,083.09 1,074–1,092 — —
Arena Score in Spanish 1,057.78 1,023–1,092 — —
Arena Score, overall 1,050.55 1,043–1,058 — —
Arena Score, programming 1,039.92 1,027–1,053 — —
Arena Score in French 1,037.69 1,003–1,073 — —
Arena Score, hard prompts 1,028.31 1,018–1,039 — —
Arena Score, creative writing 1,024.59 1,011–1,039 — —
Arena Score, instruction following 1,023.29 1,013–1,033 — —
Arena Score in Russian 1,002.96 987–1,019 — —
Arena Score, long queries 995.98 980–1,012 — —
Arena Score, multi-turn dialogue 988.95 973–1,004 — —
Arena Score, expert questions 983.5 963–1,004 — —