
by Together AI
Together AI is a cloud platform for running, fine-tuning and serving open-source AI models through an OpenAI-compatible API. It also offers dedicated endpoints and GPU clusters.
APIUsage-based
Step 1: Create an account and get an API key.
Step 2: Install the Python or TypeScript client, or use cURL against the chat completions endpoint.
Step 3: Pick a model from the models library and send your first request.
Step 4: Try models in the playground and export the code.
Step 5: Move to dedicated endpoints, batch jobs or fine-tuning as your workload grows.
Step 6: Use the Together CLI to manage endpoints, fine-tunes and GPU clusters.
Yes. The docs describe an OpenAI-compatible API for running open-source models.
Serverless charges by the tokens you use, while dedicated inference provisions single-tenant GPUs for predictable, isolated latency and charges by GPU hours.
Yes. You can upload your own weights and serve them through the Together AI API, or run your own Docker containers on managed GPUs.
It needs Python 3.10 or later and uv. Install it with uv tool install "together[cli]" and run tg --help to list commands.
API endpoint
Requirements
Environment variables
BittensorBittensor is an open network where independent subnets produce digital commodities such as compute, inference, storage and prediction, and contributors are paid in TAO. A Python SDK and the btcli command line give access to the chain.
Chainbase AIChainbase AI is a set of tools that gives AI agents access to onchain data from 90+ blockchains. It includes agent skills, a command-line tool and an MCP server.