Providers.
CSuite runs every model through one of five providers: Runware and Replicate in the cloud with your own API keys, or the Ollama, Hugging Face, and GGML runtimes fully local on your machine.
Runware is a cloud inference platform built for fast, low-cost media generation. CSuite connects to it with your own Runware API key: requests go straight from your machine to Runware, generations are billed to your account at Runware's own rates, and the outputs land as plain files on your disk.
Replicate hosts a wide range of AI models behind one API, including several frontier models the day they ship. CSuite connects to it with your own Replicate API token: you pay Replicate directly for what you run, and everything you generate stays on your machine.
Ollama is the local runtime bundled with CSuite. Open-weight language and image models are downloaded once and then run entirely on your machine — no API key, no per-use fees, and nothing leaves your computer. What you can run is gated by your hardware, and the catalog lists each model's requirements.
CSuite runs ONNX builds of open models from Hugging Face fully locally, in-process via transformers.js. Like Ollama models, they are downloaded once and run on your own hardware — no API key and no per-use fees — with smaller models that work well even on modest machines.
GGML is the local runtime CSuite bundles for on-device image and video generation, built on stable-diffusion.cpp. Model weights are downloaded once and every frame is rendered on your own hardware — no API key and no per-use fees. It's the most demanding of the local runtimes, so each model lists the machine it needs.