Mistral AI

Name: ProtectMyAPI
Rating: 4.9 (50 reviews)

Integrate Mistral Large, Codestral, and Mixtral with ProtectMyAPI.

💨

Supported Models: Mistral Large, Mistral Small, Codestral, Mixtral 8x7B

Features

✅ Chat Completions
✅ Code generation (Codestral)
✅ Fill-in-the-middle (FIM)
✅ Function Calling
✅ JSON Mode
✅ Streaming responses

Setup

1. Add your API key to Secrets

In the ProtectMyAPI dashboard:

Go to your app → Secrets
Add a secret named MISTRAL_API_KEY
Paste your Mistral API key as the value

2. Create an endpoint

Create an endpoint with:

Name: Mistral Chat
Slug: mistral-chat
Target URL: https://api.mistral.ai/v1/chat/completions
Method: POST
Auth Type: Bearer
Auth Value: {{MISTRAL_API_KEY}}

Code Examples

import ProtectMyAPI
 
// Simple chat
let response = try await ProtectMyAPI.shared.request(
    endpoint: "mistral-chat",
    method: .POST,
    body: [
        "model": "mistral-large-latest",
        "messages": [
            ["role": "user", "content": "Explain machine learning"]
        ]
    ]
)
 
// Code generation with Codestral
let codeResponse = try await ProtectMyAPI.shared.request(
    endpoint: "mistral-chat",
    method: .POST,
    body: [
        "model": "codestral-latest",
        "messages": [
            ["role": "user", "content": "Write a binary search in Python"]
        ]
    ]
)
 
// JSON Mode
let jsonResponse = try await ProtectMyAPI.shared.request(
    endpoint: "mistral-chat",
    method: .POST,
    body: [
        "model": "mistral-large-latest",
        "messages": [
            ["role": "user", "content": "List 3 programming languages with their year of creation as JSON"]
        ],
        "response_format": ["type": "json_object"]
    ]
)

val response = ProtectMyAPI.instance.request(
    endpoint = "mistral-chat",
    method = "POST",
    body = JSONObject().apply {
        put("model", "mistral-large-latest")
        put("messages", JSONArray().apply {
            put(JSONObject().apply {
                put("role", "user")
                put("content", "Explain machine learning")
            })
        })
    }
)

final response = await ProtectMyAPI.instance.secureRequest(
  '/mistral-chat',
  body: {
    'model': 'mistral-large-latest',
    'messages': [
      {'role': 'user', 'content': 'Explain machine learning'}
    ],
  },
);

Models

Model	Best For	Context
`mistral-large-latest`	Complex tasks	128K
`mistral-small-latest`	Fast, simple tasks	128K
`codestral-latest`	Code generation	32K
`open-mixtral-8x7b`	Open source	32K

Fill-in-the-Middle (FIM)

Codestral supports FIM for code completion:

Create an endpoint:

Slug: codestral-fim
Target URL: https://api.mistral.ai/v1/fim/completions

let fimResponse = try await ProtectMyAPI.shared.request(
    endpoint: "codestral-fim",
    method: .POST,
    body: [
        "model": "codestral-latest",
        "prompt": "def fibonacci(n):",
        "suffix": "    return result",
        "max_tokens": 256
    ]
)

Function Calling

let response = try await ProtectMyAPI.shared.request(
    endpoint: "mistral-chat",
    method: .POST,
    body: [
        "model": "mistral-large-latest",
        "messages": [
            ["role": "user", "content": "What's the weather in Paris?"]
        ],
        "tools": [
            [
                "type": "function",
                "function": [
                    "name": "get_weather",
                    "description": "Get weather for a location",
                    "parameters": [
                        "type": "object",
                        "properties": [
                            "location": ["type": "string"]
                        ],
                        "required": ["location"]
                    ]
                ]
            ]
        ]
    ]
)

Google Gemini Groq