Roc package for the OpenRouter.ai API

This package is still in WIP 🛠️ stages, so the interface may be subject to change. With that said, the package currently supports:

• Creating and parsing ChatML style requests and responses.
• Creating and parsing raw prompt style requests and responses.
• Formatting prompt strings with [INST], <<SYS>>, and <s> tags for models with llama style fine-tuning.
• Access to over 230 open and closed source models, including models from OpenAI, Anthropic, Mistral, Meta, Google, Microsoft, Perplexity, and more.
• Most common LLM parameters such as temperature, topP, topK, repetitionPenalty, etc.
• OpenRouter specific features like fallback models and provider preferences.
• 🚀 NEW! LLM tool use - this enables the AI model to call Roc functions and use the results in its answers.
  • Includes a collection of prebuilt tools, or you can build your own

Example

import cli.Stdout
import cli.Http
import ai.Chat

main =
    apiKey = "<your_api_key>"
    client = Chat.initClient { apiKey, model: "openai/gpt-4o" }
    messages = Chat.appendUserMessage [] "Hello, world!" {}
    response = Http.send! (Chat.buildRequest client messages)
    when Chat.decodeTopMessageChoice response.body is
        Ok message -> Stdout.line! message.content
        Err (ApiError err) -> Stdout.line! "$(Num.toStr err.code): $(err.message)"
        Err NoChoices -> Stdout.line! "No message choices in API response"
        Err (BadJson str) -> Stdout.line! "Error parsing JSON:\n$(str)"
        Err DecodingError -> Stdout.line! "Error decoding API response"

For complete example apps, including a full chatbot app with tool use, see the examples folder.