Connect your DAW to any LLM

Configuration Model

Four-source configuration merge system with defaults, config file, environment variables, and plugin UI override for musegpt.

Inference Backend

Client interface, OpenAI-compatible API translation, streaming and structured response modes, per-request metrics, and transcription support.

Inter-Process Communication Model

Ports and adapters architecture for thread-to-thread and plugin-to-backend communication in musegpt.

Threading Model

Three-thread architecture for real-time audio, responsive UI, and background inference in the musegpt VST3 plugin.

Anthropic

Cloud API provider for Claude models via the Anthropic Messages API, a non-OpenAI-compatible backend.

DeepSeek

Cloud API provider for DeepSeek chat and reasoning models via api.deepseek.com.

Fireworks AI

Cloud API provider for fast open-model inference via api.fireworks.ai.

Google Gemini

Cloud API provider for Gemini models via the OpenAI-compatible endpoint at generativelanguage.googleapis.com.

Groq

Cloud API provider for fast inference on Groq LPU hardware via api.groq.com.

llama.cpp

High-performance local inference engine with OpenAI-compatible server mode.

Mistral AI

Cloud API provider for Mistral models via api.mistral.ai.

Ollama

Local model runner with an OpenAI-compatible API. The default backend for musegpt.

OpenAI

Cloud API provider for GPT-4o, o1, and other OpenAI models via api.openai.com.

OpenRouter

Cloud API aggregator providing access to many models from multiple providers through a single OpenAI-compatible endpoint.

Together AI

Cloud API provider popular for open-weight models via api.together.xyz.

vLLM

High-throughput LLM serving engine with OpenAI-compatible API.

whisper.cpp

Local speech recognition engine for audio-to-text transcription in the musegpt audio pipeline.