Mistral Medium 3.5 pricing & specs

Mistral · released 2026-04-29 · ● verified 2026-10-11

input / 1M$1.50
output / 1M$7.50
context262.1K
max output262.1K

// price history ($/1M tokens)

$8.625
input $/1Moutput $/1M2026-04-29 → 2026-10-11

// cheaper or similar alternatives

modelinoutctxvs Mistral Medium 3.5
Gemma 4 E2B IT$0.04$0.08131.1K−98.3%
Qwen3.7 Flash$0.03$0.131M−98.2%
Qwen3.5 9B$0.10$0.15262.1K−96.2%
Gemma 4 26B A4B IT$0.07$0.34262.1K−95.4%
GPT-5 Nano$0.05$0.40400K−95.4%
Claude Haiku 4.5 vs Mistral Medium 3.5 →Gemini 3.7 Flash vs Mistral Medium 3.5 →Gemini 3.8 Flash vs Mistral Medium 3.5 →Mistral Medium 3.5 vs Muse Spark 1.2 →Mistral Medium 3.5 vs Muse Spark 1.3 →GPT-5.4 mini vs Mistral Medium 3.5 →

// price by host

hostin $/1Mout $/1M
Scaleway$1.50$7.50

// faq

How much does Mistral Medium 3.5 cost?

Mistral Medium 3.5 costs $1.50 per million input tokens and $7.50 per million output tokens. At 50M input and 10M output tokens per month that is about $150.

What is the context window of Mistral Medium 3.5?

262.1K tokens, with up to 262.1K output tokens per request.

Does Mistral Medium 3.5 support prompt caching?

Yes. Cached input tokens cost $0.15 per million.

What are cheaper alternatives to Mistral Medium 3.5?

Gemma 4 E2B IT ($0.04 in / $0.08 out), Qwen3.7 Flash ($0.03 in / $0.13 out), Qwen3.5 9B ($0.10 in / $0.15 out) offer the same input types and capabilities for less.

Is Mistral Medium 3.5 open weights?

Yes. It is available from 1 tracked hosts; the cheapest is Scaleway at $1.50 input and $7.50 output per million tokens.

// spec sheet

input / 1M
$1.50
output / 1M
$7.50
cache read / 1M
$0.15
blended (3:1)
$3.00
context
262.1K
max output
262.1K
input types
text · image
reasoning
yes
tool calling
yes
structured output
yes
open weights
yes
knowledge cutoff
n/a
price source
Mistral

report a data error

// cost calculator

modelin $/1Mout $/1M= monthly
Gemma 4 E2B IT$0.04$0.08$2.80
Qwen3.7 Flash$0.03$0.13$2.80
Qwen3.5 9B$0.10$0.15$6.50
Mistral Medium 3.5$1.50$7.50$150

Base-tier list prices. Long-context tiers, caching and batch discounts are not included.