Mistral AI

Mixtral 8x7B

Type
language model
Context
33K tokens
Max output
33K
Released
11 December 2025
API string
mistralai/mixtral-8x7b-instruct-v0.1
Weights
open

Prices by provider

Provider Input, $ per 1M Output, $ per 1M In our data since
anyscale $0.15 $0.15 31 Jul 2026
NanoGPT $0.27 $0.27 31 Jul 2026
Replicate $0.3 $1 31 Jul 2026
DeepInfra $0.4 $0.4 31 Jul 2026
Cortecs $0.438 $0.68 31 Jul 2026
Together AI $0.6 $0.6 31 Jul 2026
OVHCloud AI Endpoints $0.63 $0.63 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 in English 1,167.7 1,162–1,173
Arena Score in French 1,164.92 1,142–1,188
Arena Score, mathematics 1,147.16 1,139–1,156
Arena Score, overall 1,131.72 1,127–1,136
Arena Score, programming 1,126.86 1,119–1,135
Arena Score, hard prompts 1,114.99 1,108–1,122
Arena Score, multi-turn dialogue 1,114.65 1,106–1,123
Arena Score in Spanish 1,110.19 1,086–1,134
Arena Score, instruction following 1,109.85 1,104–1,116
Arena Score, creative writing 1,109.18 1,101–1,118
Arena Score, long queries 1,103.07 1,094–1,113
Arena Score in Russian 1,089.45 1,079–1,100
Arena Score, expert questions 1,087.25 1,074–1,100
MATH, difficulty level five 9.95 % · with a tuned harness Epoch evaluations
GPQA Diamond — graduate-level questions 7.45 % · with a tuned harness Epoch evaluations