MistralSharp 0.1.0

MistralSharp

About

MistralSharp is a .NET client library for interfacing with the Mistral AI platform written in C# using .NET 8. Features async support and no external dependencies.

Usage

Check out the Sample project to see an example of usage.

Start by downloading the nuget package and importing it into your project. Then, creating a new instance of MistralClient and pass your API key:

var mistralClient = new MistralClient(apiKey);

Endpoints

To return a list of available models:

var models = await mistralClient.GetAvailableModelsAsync();

To chat with a model, first create a new ChatRequest object (note: only Model and Messages must be supplied, the other parameters are optional and will default to the values specified below:

var chatRequest = new ChatRequest()
{
    
    // The ID of the model to use. You can use GetAvailableModelsAsync() to get the list of available models
    Model = "mistral-medium",
    
    // Pass a list of messages to the model. 
    // The role can either be "user" or "agent"
    // Content is the message content
    Messages =
    [
        new Message()
        {
            Role = "user",
            Content = "How can Mistral AI assist programmers?"
        }
    ],
    
    //The maximum number of tokens to generate in the completion.
    // The token count of your prompt plus max_tokens cannot exceed the model's context length.
    MaxTokens = 16,
    
    //  Default: 0.7
    // What sampling temperature to use, between 0.0 and 2.0.
    // Higher values like 0.8 will make the output more random, while lower values like 0.2 will make
    // it more focused and deterministic.
    Temperature = 0.7,
    
    //  Default: 1
    // Nucleus sampling, where the model considers the results of the tokens with top_p probability mass.
    // So 0.1 means only the tokens comprising the top 10% probability mass are considered.
    // Mistral generally recommends altering this or temperature but not both.
    TopP = 1,
    
    //  Default: false
    // Whether to stream back partial progress. If set, tokens will be sent as data-only server-sent events
    // as they become available, with the stream terminated by a data: [DONE] message. Otherwise, the server will
    // hold the request open until the timeout or until completion, with the response containing the full
    // result as JSON.
    Stream = false,
    
    //  Default: false
    // Whether to inject a safety prompt before all conversations.
    SafeMode = false,
    
    //  Default: null
    // The seed to use for random sampling. If set, different calls will generate deterministic results.
    RandomSeed = null
};

Finally, call the ChatAsync() method and pass the ChatRequest object:

var sampleChat = await mistralClient.ChatAsync(chatRequest);

Note

• More in-depth documentation coming soon!
• This project is currently in active development so expect frequent changes in the coming weeks.