M

Mistral AI (Mistral Large / Small / Codestral)

LLM

by Mistral AI

French AI company offering efficient open and commercial models. Mistral Large for complex reasoning, Mistral Small for cost-effective tasks, and Codestral for code generation. Known for strong European data privacy.

Registration & API Key Steps

1

Step 1

Visit console.mistral.ai and sign up with email, Google, or GitHub.

Open link
2

Step 2

Verify your phone number to activate the free Experiment plan.

3

Step 3

Choose between Experiment (free) or Scale (pay-as-you-go) plan.

4

Step 4

Navigate to API Keys section and click "Create new key".

5

Step 5

Copy and securely store your API key.

Pricing

TierPriceFeatures
Mistral Nemo$0.02 / $0.06 per 1M tokensInput / Output. Most affordable. Open-source.
Mistral Small$0.10 / $0.30 per 1M tokensInput / Output. Good for simple tasks.
Mistral Medium 3$0.40 / $2.00 per 1M tokensInput / Output. Balanced performance.
Mistral Large$2.00 / $6.00 per 1M tokensInput / Output. Most capable for complex reasoning.
Codestral$0.30 / $0.90 per 1M tokensInput / Output. Specialized for code generation.

Application Tips

Tip 1

Experiment plan is free — great for prototyping without credit card commitment.

Tip 2

Mistral Nemo at $0.02/1M input tokens is one of the cheapest commercial APIs.

Tip 3

Codestral is specifically optimized for code — use it for programming tasks.

Tip 4

EU-based servers ensure GDPR compliance, important for European customers.

Tip 5

API is OpenAI-compatible, making migration from OpenAI easy.

Tip 6

Open-source models (Mistral Nemo, Mixtral) can be self-hosted for free.

China Access Solutions

Access Solution

Requires VPN/proxy for API access. Consider self-hosting open-source Mistral models (Nemo, Mixtral) in China for local access.

Code Example

JavaScript / TypeScript
from mistralai import Mistral

client = Mistral(api_key="your-api-key")

response = client.chat.complete(
    model="mistral-large-latest",
    messages=[
        {"role": "user", "content": "What is the best French cheese?"}
    ]
)

print(response.choices[0].message.content)

# --- cURL example ---
# curl https://api.mistral.ai/v1/chat/completions \
#   -H "Authorization: Bearer your-api-key" \
#   -H "Content-Type: application/json" \
#   -d '{
#     "model": "mistral-large-latest",
#     "messages": [{"role": "user", "content": "Hello!"}]
#   }'

Rate Limits

TierLimits
DefaultExperiment (free): 1 RPM per model. Scale: varies by model, typically 5-50 RPM. Enterprise: custom limits.

Recommended Use Cases

Multilingual tasks (especially European languages)Code generationGDPR-compliant applicationsCost-effective chatbotsResearch
Last Updated: 2026-02-10

Related API Guides