Mistral AI Deploy Integration

Overview

Deploy Mistral AI-powered applications to production with proper API key management, model endpoint configuration, and platform-specific optimizations. Covers Vercel, Docker, and Cloud Run deployments with the Mistral SDK connecting to api.mistral.ai.

Prerequisites

• Mistral AI API key for production use
• Platform CLI installed (vercel, docker, or gcloud)
• Application using @mistralai/mistralai SDK
• Environment variables documented

Instructions

Step 1: Configure Environment Variables

# Vercel
vercel env add MISTRAL_API_KEY production
vercel env add MISTRAL_MODEL production  # e.g., mistral-large-latest

# Docker
echo "MISTRAL_API_KEY=your-key" > .env.production

# Cloud Run
gcloud secrets create mistral-api-key --data-file=- <<< "your-key"

Step 2: Docker Deployment

FROM node:20-slim AS builder
WORKDIR /app
COPY package*.json ./
RUN npm ci
COPY . .
RUN npm run build

FROM node:20-slim
WORKDIR /app
COPY --from=builder /app/dist ./dist
COPY --from=builder /app/node_modules ./node_modules
COPY --from=builder /app/package.json ./

ENV NODE_ENV=production
EXPOSE 3000  # 3000: 3 seconds in ms
CMD ["node", "dist/index.js"]

Step 3: Vercel Serverless Configuration

{
  "functions": {
    "api/chat.ts": {
      "maxDuration": 60
    }
  },
  "env": {
    "MISTRAL_API_KEY": "@mistral_api_key"
  }
}

// api/chat.ts - Vercel Edge Function
import Mistral from "@mistralai/mistralai";

export const config = { runtime: "edge" };

export default async function handler(req: Request) {
  const client = new Mistral({ apiKey: process.env.MISTRAL_API_KEY! });
  const { messages } = await req.json();

  const response = await client.chat.complete({
    model: process.env.MISTRAL_MODEL || "mistral-small-latest",
    messages,
  });

  return Response.json(response);
}

Step 4: Cloud Run Deployment

set -euo pipefail
gcloud run deploy mistral-service \
  --image gcr.io/$PROJECT_ID/mistral-app \
  --region us-central1 \
  --platform managed \
  --set-secrets=MISTRAL_API_KEY=mistral-api-key:latest \
  --set-env-vars=MISTRAL_MODEL=mistral-large-latest \
  --min-instances=1 \
  --max-instances=10

Step 5: Health Check Endpoint

import Mistral from "@mistralai/mistralai";

export async function GET() {
  try {
    const client = new Mistral({ apiKey: process.env.MISTRAL_API_KEY! });
    await client.models.list();
    return Response.json({ status: "healthy", provider: "mistral" });
  } catch (error) {
    return Response.json({ status: "unhealthy", error: error.message }, { status: 503 });  # HTTP 503 Service Unavailable
  }
}

Error Handling

Issue	Cause	Solution
API key not found	Missing env var	Verify secret configuration on platform
Function timeout	Long completion	Increase maxDuration, use streaming
Cold start latency	Serverless spin-up	Set minimum instances or use edge runtime
Model not available	Wrong model ID	Check available models at console.mistral.ai

Examples

Production Client Wrapper

const client = new Mistral({
  apiKey: process.env.MISTRAL_API_KEY!,
  timeout: 60000,  # 60000: 1 minute in ms
  maxRetries: 3,
});

Resources

• Mistral AI API Documentation
• Mistral AI Models

Next Steps

For multi-environment setup, see mistral-multi-env-setup.

Output

• Configuration files or code changes applied to the project
• Validation report confirming correct implementation
• Summary of changes made and their rationale