Mistral API pricing for every model, newest first
By TechCompare · Updated · How we check
A 30,000 token prompt with a 2,000 token answer and a quarter of the input cached costs $0.020475 on Mistral Large 4, the newest model. Ministral 3 3B 2512, the lowest cost model Mistral sells itself, runs it for $0.002525.
Every Mistral model and its API price
Newest first. Prices are per 1M tokens. One request is 30K tokens in and 2K out, with 25% of the input cached.
| Model | Input | Output | Cached | Context | One request | Listed |
|---|---|---|---|---|---|---|
| Mistral Large 4Mistral | $0.70 | $2.10 | $0.07 | 1.05M | $0.0205 | Oct 6, 2026 |
| Mistral Medium 3.5Mistral | $1.50 | $7.50 | $1.50 | 262K | $0.06 | Apr 30, 2026 |
| Mistral Small 4Mistral | $0.15 | $0.60 | $0.015 | 262K | $0.0047 | Mar 16, 2026 |
| Devstral 2 2512Mistral | $0.40 | $2 | $0.04 | 262K | $0.0133 | Dec 9, 2025 |
| Ministral 3 14B 2512Mistral | $0.20 | $0.20 | $0.02 | 262K | $0.005 | Dec 2, 2025 |
| Ministral 3 3B 2512Mistral | $0.10 | $0.10 | $0.01 | 131K | $0.0025 | Dec 2, 2025 |
| Ministral 3 8B 2512Mistral | $0.15 | $0.15 | $0.015 | 262K | $0.0038 | Dec 2, 2025 |
| Mistral Large 3 2512Mistral | $0.50 | $1.50 | $0.05 | 262K | $0.0146 | Dec 1, 2025 |
| Voxtral Small 24B 2507Mistral | $0.10 | $0.30 | $0.01 | 33K | $0.0029 | Oct 30, 2025 |
| Mistral Medium 3.1Mistral | $0.40 | $2 | $0.04 | 131K | $0.0133 | Aug 13, 2025 |
| Codestral 2508Mistral | $0.30 | $0.90 | $0.03 | 256K | $0.0088 | Aug 1, 2025 |
| Mistral Small 3.2 24BMistral | $0.10 | $0.30 | $0.01 | 256K | $0.0029 | Jun 20, 2025 |
Listed is the day OpenRouter added the model. Prices checked Nov 14, 2026.
Mistral AI is a lab in Paris that sells its models through its own API. This page lists all 20 Mistral models OpenRouter carries, newest first, at Mistral's own price where Mistral sells the model and the host that charges least where it doesn't.
The general models come in three sizes. Mistral Large 4 is the frontier model with a 1.05M window, at $0.70 input and $2.10 output per million tokens. Mistral Medium 3.5 is a dense 128B model, at $1.50 and $7.50. Mistral Small 4 pulls several of Mistral's flagship skills into one small model, at $0.15 and $0.60.
Calculator
Cost Comparison
Based on 30,000 input tokens (25% cached), 2,000 output tokens, and 100 requests.Prices come live from OpenRouter at each maker's own price where it's listed. Temporary deals are left out.
How to read the Mistral prices
Mistral's price list doesn't follow model size. Mistral Large 4 costs $0.70 per million input tokens, below Mistral Medium 3.5 at $1.50, so check both before you assume the big model costs more.
Mistral also sells models built for one job each. Devstral 2 2512 is a 123B open source model for agent coding, at $0.40 input and $2 output per million. Codestral handles quick code completion, including fill in the middle. Voxtral works with audio, Saba is trained on the languages of the Middle East and South Asia, and the Ministral models come in 3B, 8B and 14B sizes for edge and on device use.
The Batch API takes 50% off input on Mistral Medium 3.5, for $0.75 input and $3.75 output per million. A cached token on Mistral Large 4 costs $0.07 per million, 90% off the normal input rate.
Mistral publishes open weights for many of its models, so hosts sell them and you can run them yourself. Mistral Large 4 reads up to 1.05M tokens, the largest window in the family.
Verdict
Start with Mistral Large 4 and run Ministral 3 3B 2512, the lowest cost model Mistral sells for the example request, on the same prompts. Small models do well on narrow jobs like tagging and extraction, so try it there first.
For coding, test Devstral 2 2512 before a general model. Send anything that can wait through the Batch API.
More LLM pricing scenarios
Related guides
Frequently asked questions
How much does the Mistral API cost?
Which Mistral model is best for coding?
What do Ministral, Codestral, Voxtral and Saba mean?
Is there a batch discount on the Mistral API?
Can I run Mistral models myself?
Which Mistral model has the largest context window?
Where do these prices come from?
Related tools
LLM VRAM Calculator
Calculate the VRAM needed to run or fine-tune any LLM at any quantization.
Use toolPower Cost Estimator
Estimate annual electricity costs for your PC, Server, or TV.
Use toolData Transfer Calculator
Estimate transfer times for files over USB, WiFi, Ethernet, and more.
Use toolJSON Formatter
Validate, format, and minify JSON. Browse it as a tree, run JSONPath queries, compare two documents, or validate against a JSON Schema.
Use tool