API pricing

Mistral AI Pricing

Mistral AI charges between $0.040 and $7.50 per 1 million output tokens, depending on the model. Ministral 3B is the cheapest at $0.040/1M output and $0.040/1M input; Mistral Medium is the most expensive at $7.50/1M output. The largest context window is 262K tokens (Devstral 2). Prices are USD, current as of July 20, 2026.

Mistral spans open-weight models you can self-host and hosted-only commercial models. The hosted rates below are what you pay through the Mistral API — self-hosting shifts the cost to your own GPU spend, which the calculator can model separately.

Last updated · synced weekly from the upstream model catalog

Cheapest input

$0.020/1M

Mistral Nemo

Cheapest output

$0.040/1M

Ministral 3B

Longest context

262K

Devstral 2

Models priced

85

Avg $2.25/1M output

Pricing by model

All prices in USD per 1 million tokens. Showing 80 of 85 priced models, cheapest first.

ModelInput / 1MOutput / 1MContext
Ministral 3B$0.040$0.040131K
Mistral Nemo$0.020$0.040131K
Mistral 7B Instruct$0.028$0.05433K
Mistral Small 3$0.050$0.08033K
Ministral 3 3B 2512$0.100$0.100131K
Ministral 8B$0.100$0.100131K
Pixtral 12B$0.100$0.10033K
Devstral Small 2505$0.060$0.120128K
Ministral 3 8B 2512$0.150$0.150262K
Ministral 8B 2512$0.150$0.150262K
Ministral 8B Latest$0.150$0.150262K
Open Mistral Nemo
Tools
$0.150$0.150128K
Pixtral 12B 2409$0.150$0.150128K
Mistral Small 3.2 24B$0.060$0.180131K
Mistral Small 3.2 2506$0.060$0.180131K
Mistral Small Latest$0.060$0.180131K
Mistral 7B Instruct v0.1$0.110$0.1903K
Ministral 3 14B 2512$0.200$0.200262K
Mistral 7B Instruct v0.2$0.200$0.20033K
Mistral 7B Instruct v0.3$0.200$0.20033K
Mistral Small 3.1 24B$0.050$0.220131K
Codestral Mamba Latest$0.250$0.250256K
Mistral 7B
Tools
$0.250$0.2508K
Mistral Tiny$0.250$0.25033K
Open Codestral Mamba$0.250$0.250256K
Open Mistral 7B$0.250$0.25032K
Devstral Small 1.1$0.070$0.280128K
Devstral Small
Tools
$0.100$0.300128K
Devstral Small 2507$0.100$0.300128K
Devstral Small Latest$0.100$0.300256K
Labs Devstral Small 2512$0.100$0.300256K
Mistral Small 3.2
VisionTools
$0.100$0.300128K
Open Mistral Nemo 2407$0.300$0.300128K
Voxtral Small 24B 2507$0.100$0.30032K
Mixtral 8x7B Instruct$0.540$0.54033K
Mistral Small$0.200$0.60033K
Mistral Small 4
VisionReasoningTools
$0.150$0.600256K
Saba$0.200$0.60033K
Mixtral 8x7B
Tools
$0.700$0.70032K
Open Mixtral 8X7B$0.700$0.70032K
Codestral
Tools
$0.300$0.900256K
Codestral 2501$0.300$0.900256K
Codestral 2508$0.300$0.900256K
Magistral Small
ReasoningTools
$0.500$1.50128K
Magistral Small 1.2 2509$0.500$1.5040K
Magistral Small 2506$0.500$1.5040K
Magistral Small Latest$0.500$1.5040K
Mistral Large 2512$0.500$1.50262K
Mistral Large 3
VisionTools
$0.500$1.50262K
Mistral Large Latest$0.500$1.50262K
Devstral 2
Tools
$0.400$2.00262K
Devstral 2512$0.400$2.00256K
Devstral Latest$0.400$2.00256K
Devstral Medium$0.400$2.00131K
Devstral Medium 2507$0.400$2.00128K
Devstral Medium Latest$0.400$2.00256K
Mistral Medium 2505$0.400$2.00131K
Mistral Medium 2508$0.400$2.00131K
Mistral Medium 3$0.400$2.00131K
Mistral Medium 3.1$0.400$2.00131K
Mistral Medium 3.1 2508$0.400$2.00131K
Codestral 2405$1.00$3.0032K
Codestral Latest$1.00$3.0032K
Magistral Medium
ReasoningTools
$2.00$5.00128K
Magistral Medium 1.2 2509$2.00$5.0040K
Magistral Medium 2506$2.00$5.0041K
Magistral Medium 2506 (thinking)$2.00$5.0041K
Magistral Medium 2509$2.00$5.0040K
Magistral Medium Latest$2.00$5.0040K
Mistral Large$2.00$6.00128K
Mistral Large 2.1
Tools
$2.00$6.00131K
Mistral Large 2407$2.00$6.00131K
Mistral Large 2411$2.00$6.00131K
Mixtral 8x22B
Tools
$2.00$6.0064K
Mixtral 8x22B Instruct$2.00$6.0066K
Open Mixtral 8X22B$2.00$6.0065K
Pixtral Large
VisionTools
$2.00$6.00128K
Pixtral Large 2411$2.00$6.00131K
Pixtral Large Latest$2.00$6.00128K
Mistral Large 2402$4.00$12.0032K

Frequently asked questions

How much does Mistral AI cost per 1M tokens?

Mistral AI pricing ranges from $0.040 to $7.50 per 1 million output tokens across 85 models. Input tokens are cheaper than output on every model. Rates current as of July 20, 2026.

What is the cheapest Mistral AI model?

Ministral 3B is the cheapest Mistral AI model on output tokens at $0.040 per 1M, while Mistral Nemo is cheapest on input at $0.020 per 1M. Which wins for you depends on your input-to-output ratio.

What is the largest Mistral AI context window?

Devstral 2 has the largest context window in the Mistral AI catalog at 262K input tokens, with up to 262K output tokens per response.

Which Mistral AI models support reasoning or vision?

The Mistral AI catalog includes 5 reasoning models, 6 vision models, 16 with tool calling. Reasoning models bill their internal thinking as output tokens, so they cost more per visible response than the headline rate suggests.

How do I calculate my actual Mistral AI bill?

Multiply your input tokens by the input rate and your output tokens by the output rate, both per million, then add retries and any conversation history resent on each turn. Calcaas does this against your real usage pattern and shows the margin left at your price point.

Head-to-head comparisons

Other providers

What will this actually cost you?

Per-token rates only tell you half the story. Calcaas multiplies them by your real usage — prompt size, response length, retries and conversation history — then shows the margin left at your price point.

Open the free cost calculator