modelbenchmark.io

Mistral Medium 3.5 Thinking

Mistral · mistral-medium-3-5

Compare

Mistral Medium 3.5 with reasoning enabled by default (reasoning_effort=high), for complex coding, agentic, and multi-step reasoning prompts.

Specification

mistral listing

Context
262K
Max output
262K
Released
2026-04-30
Knowledge cutoff
Retires
Open weights
yes
Input
text, image
Output
text

Price

US dollars per million tokens · mistral list price

Input
$1.50
Output
$7.50
Cache read
$0.15
Cache write
$1.50

Available from 10 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
mistralvendor$1.50$7.50$0.15262K262K
kenari · freefreefree262K262K
kilo$1.50$7.50262K210K
nano-gpt$1.50$7.50$0.75256K33K
nano-gpt · thinking$1.50$7.50$0.75256K33K
openrouter$1.50$7.50262K210K
pioneer$1.50$7.50$1.50$1.50256K131K
vercel$1.50$7.50256K256K
cortecs$1.532$7.843$0.154256K256K
requesty$1.65$8.25$1.65262K262K

Price history

append-only observations · a listing writes a row only when its price moves

DateHostField$/M
2026-09-02cortecsCache read$0.139
2026-09-15cortecsCache read$0.154
2026-09-02cortecsInput$1.393
2026-09-15cortecsInput$1.532
2026-09-02openrouterInput$1.50
2026-09-06openrouterInput$1.50
2026-09-02cortecsOutput$7.13
2026-09-15cortecsOutput$7.843
2026-09-02openrouterOutput$7.50
2026-09-06openrouterOutput$7.50

More from Mistral

most-hosted first