AI
Run and serve open large language models
Run open large language models on remote compute — served to your browser with a chat interface and an HTTP API for your own apps.
This category covers:
- Local LLMs — run open models (Llama, Qwen, Gemma, and more) with Ollama, GPU-accelerated when available
- Managed access — accounts, groups and per-model permissions in front of those models, so a team shares one GPU with a chat interface and an API key each, from a step that either carries the model server or fronts the machines that do
Browse the workflows below. To deploy one, see How it works.