Mistral AI Setup

Mistral is available in AxonFlow Community and Enterprise. As a leading European AI company based in France, Mistral offers high-performance models with competitive pricing and EU data residency by default.

Runtime Defaults

AxonFlow's Mistral provider defaults to:

• Provider name: mistral
• Default model: mistral-small-latest
• Default endpoint: https://api.mistral.ai

Environment Variables

export MISTRAL_API_KEY=your-mistral-api-key
export MISTRAL_MODEL=mistral-small-latest
export MISTRAL_TIMEOUT_SECONDS=120           # Request timeout in seconds (default: 120)
export MISTRAL_ENDPOINT=https://api.mistral.ai  # Custom endpoint (e.g., self-hosted)

YAML Configuration

version: "1.0"

llm_providers:
  mistral:
    enabled: true
    credentials:
      api_key: ${MISTRAL_API_KEY}
    config:
      model: ${MISTRAL_MODEL:-mistral-small-latest}

Models

AxonFlow accepts any model alias that the Mistral API supports. The -latest aliases automatically resolve to the newest version.

Model alias	Tier	AxonFlow default
mistral-small-latest	Small (fast, cost-effective)	✅
mistral-medium-latest	Medium (balanced)
mistral-large-latest	Large (most capable)
codestral-latest	Code generation
ministral-8b-latest	Lightweight, low latency

Mistral updates model versions and pricing frequently. For the current model list, pricing, and context windows, see the Mistral models documentation.

tip

Use -latest aliases (e.g., mistral-small-latest) rather than pinned version IDs unless you need reproducibility. The aliases automatically pick up new model releases without config changes.

Proxy Mode Example

cURL

curl -X POST http://localhost:8080/api/request \
  -H "Content-Type: application/json" \
  -d '{
    "query": "Explain the EU AI Act in two sentences.",
    "context": {
      "provider": "mistral"
    }
  }'

Python

from axonflow import AxonFlow  # v6.0.0+

async with AxonFlow(
    endpoint="http://localhost:8080",
    client_id="community",
) as client:
    response = await client.proxy_llm_call(
        query="Explain the EU AI Act in two sentences.",
        context={"provider": "mistral"},
    )
    print(response.data)

Go

import axonflow "github.com/getaxonflow/axonflow-sdk-go/v5"

client := axonflow.NewClient(axonflow.AxonFlowConfig{
    Endpoint: "http://localhost:8080",
    ClientID: "community",
})

resp, _ := client.ProxyLLMCall(axonflow.ProxyLLMCallRequest{
    Query:   "Explain the EU AI Act in two sentences.",
    Context: map[string]interface{}{"provider": "mistral"},
})
fmt.Println(resp.Data)

TypeScript

import { AxonFlow } from '@axonflow/sdk';  // v5.0.0+

const client = new AxonFlow({
  endpoint: 'http://localhost:8080',
  clientId: 'community',
});

const resp = await client.proxyLLMCall({
  query: 'Explain the EU AI Act in two sentences.',
  context: { provider: 'mistral' },
});
console.log(resp.data);

Gateway Mode Example

Gateway mode lets you call Mistral directly while AxonFlow handles policy evaluation and audit logging.

# 1. Pre-check with AxonFlow
PRECHECK=$(curl -s -X POST http://localhost:8080/api/policy/pre-check \
  -H "Content-Type: application/json" \
  -d '{"client_id": "community", "query": "Analyze customer data"}')
CONTEXT_ID=$(echo "$PRECHECK" | jq -r '.context_id')

# 2. Call Mistral directly
RESPONSE=$(curl -s https://api.mistral.ai/v1/chat/completions \
  -H "Authorization: Bearer $MISTRAL_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "mistral-small-latest",
    "messages": [{"role": "user", "content": "Analyze customer data"}],
    "max_tokens": 500
  }')

# 3. Audit the call
curl -X POST http://localhost:8080/api/audit/llm-call \
  -H "Content-Type: application/json" \
  -d "{
    \"client_id\": \"community\",
    \"context_id\": \"$CONTEXT_ID\",
    \"provider\": \"mistral\",
    \"model\": \"mistral-small-latest\",
    \"latency_ms\": 500,
    \"token_usage\": {\"prompt_tokens\": 10, \"completion_tokens\": 50, \"total_tokens\": 60}
  }"

Multi-Provider Routing

Use Mistral alongside other providers with weighted routing:

export MISTRAL_API_KEY=...
export OPENAI_API_KEY=...
export LLM_ROUTING_STRATEGY=weighted

AxonFlow distributes requests across healthy providers based on configured weights. See Provider Routing for details.

Current Capabilities

• Chat completions (text-in, text-out)
• Streaming (SSE)
• Code generation (via Codestral)
• Cost estimation
• Health monitoring

Planned (not yet implemented):

• Function calling / tool use
• Vision (Pixtral multimodal)
• JSON mode (response_format)

Getting an API Key

1. Go to console.mistral.ai
2. Sign up or sign in
3. Navigate to API Keys in the left sidebar
4. Click Create new key
5. Copy the key and set it as MISTRAL_API_KEY

Troubleshooting

Error	Cause	Solution
401 Unauthorized	Invalid or expired API key	Regenerate at console.mistral.ai
429 Too Many Requests	Rate limit exceeded	Reduce request frequency or upgrade Mistral plan
Provider not in health check	MISTRAL_API_KEY not set	Set the environment variable before starting AxonFlow
Provider limit exceeded	Community mode limits to 2 providers	Use evaluation license or set LLM_PROVIDERS=mistral to prioritize
Timeout errors	Model overloaded	Increase MISTRAL_TIMEOUT_SECONDS or use mistral-small-latest

Related Resources

• Provider Routing — Weighted, failover, and round-robin strategies
• Proxy Mode — Full proxy mode documentation
• Gateway Mode — Gateway mode documentation
• Getting Started — Quick start guide