Mistral

Set up Coreply with Mistral, including support for the low-latency FIM API with Codestral.

Coreply supports two modes with Mistral’s API: FIM (Fill-in-the-Middle) via Codestral, and standard chat completion for other Mistral models. FIM mode is more efficient and is the recommended option when using Mistral.

Mistral offers a free tier, but compatibility details with Coreply’s usage patterns are not yet fully confirmed. Check Mistral’s official documentation for current free tier limits.

FIM mode (Codestral)

FIM (Fill-in-the-Middle) is a specialized inference mode supported by Codestral. It uses fewer tokens and has lower latency than standard chat completion, making it well-suited for real-time suggestions.

FIM mode is the most efficient way to use Mistral with Coreply. Use it whenever you want the fastest response times.

1. Get your API key

   Go to console.mistral.ai/api-keys and create a new API key.

2. Enter your credentials in Coreply

   Open Coreply’s settings and fill in the following values:

   Field	Value
   API URL	https://api.mistral.ai/v1/fim/
   API Key	Your key from Mistral
   Model Name	codestral-latest

Chat completion mode

For other Mistral models, use the standard chat completion endpoint.

1. Get your API key

   Go to console.mistral.ai/api-keys and create a new API key.

2. Enter your credentials in Coreply

   Open Coreply’s settings and fill in the following values:

   Field	Value
   API URL	https://api.mistral.ai/v1/
   API Key	Your key from Mistral
   Model Name	See recommended models below

3. Choose a model

   The following models are recommended for chat completion mode:

   mistral-medium-2508
   mistral-small-2409
   

   You’re welcome to try other models from Mistral’s catalog.