This page documents AI Studio, our development environment for notebooks, jobs and deployments. Saturn Cloud also offers dedicated inference endpoints built with NVIDIA DSX. Talk to us about dedicated GPU capacity.

NLP and LLM

Templates for NLP pipelines, LLM inference, fine-tuning, and RAG

Use these templates to bootstrap language workflows from baseline inference to RAG.

LLM Inference vLLM 7B

Run a 7B instruction model with vLLM for efficient token throughput

LoRA Fine-tune

Adapter-based fine-tuning workflow for parameter-efficient LLM training

RAG Mini Docs Q&A

Minimal retrieval-augmented generation pipeline for document question answering