Skip to content
NewTry RouterPlus with free creditsTry playground →
RouterPlus
Book a demo

Mistral AI API

Open in consoleI need inference
  • 9Models
  • 2Labs
  • 1.05MMax context
  • $0.10/ $0.10From, per 1M in / out

Pricing

ModelLabContextInput / 1MCached / 1MOutput / 1MProviders
Mistral Large 4mistral-large-4Mistral AI1.05M$0.70$0.07$2.101
GLM 5.3z-ai/glm-5.3Z.ai1.05M$1.40$0.14$4.4012
Mistral Medium 3.5mistral-medium-2604Mistral AI262K$1.50$0.15$7.501
Mistral Small 4mistral-small-2603Mistral AI262K$0.15$0.015$0.601
Ministral 3 14Bministral-14b-2512Mistral AI262K$0.20$0.02$0.201
Ministral 3 3Bministral-3b-2512Mistral AI131K$0.10$0.01$0.101
Ministral 3 8Bministral-8b-2512Mistral AI262K$0.15$0.015$0.151
Mistral Large 3mistral-large-2512Mistral AI262K$0.50$0.05$1.501
Codestralcodestral-2508Mistral AI256K$0.30$0.03$0.901

Mistral AI's price, USD, as of November 5, 2026. Live prices

Quickstart

OpenAI-compatible
curl https://api.routerplus.com/v1/chat/completions \
  -H "Authorization: Bearer $ROUTERPLUS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "z-ai/glm-5.3",
    "messages": [{ "role": "user", "content": "Hello" }],
    "max_tokens": 1024,
    "provider": { "only": ["mistral-ai"] }
  }'

FAQ

How do I send a request to Mistral AI?

Routing tries providers cheapest first, by input plus output price, so Mistral AI serves a model where it is the cheapest and otherwise as a fallback. To send a request to Mistral AI first, name it in provider.only or provider.order as "mistral-ai".

How much does Mistral AI cost on RouterPlus?

You pay Mistral AI's own price for the model. Mistral AI's lowest input price is $0.10 per million tokens (Ministral 3 3B) and its lowest output price is $0.10 per million tokens (Ministral 3 3B). Prices as of November 5, 2026.

Which models does Mistral AI serve?

Mistral Large 4, GLM 5.3, Mistral Medium 3.5, Mistral Small 4, Ministral 3 14B, Ministral 3 3B, Ministral 3 8B, Mistral Large 3 and Codestral.

Does Mistral AI keep my prompts?

Mistral AI declares that it retains prompts: it keeps API prompts and answers for 30 days to monitor abuse, unless zero data retention is on for the account.

Which models does Mistral AI serve here?

Its own models (Mistral Large, Medium, Small, Ministral and Codestral), and GLM 5.3 as that model's last route.

Route to Mistral AI with one key

Open in consoleI need inference