Connect Models

Get an API key: Together AI

The cure for Together AI calls rejected for speed is a switch inside Big-AGI - Rate Limiter - which spaces Big-AGI's own requests about a second apart.

Together AI Supported Models ->
Browse documentation

When Together AI starts refusing calls for speed, the control that fixes it sits inside Big-AGI. It is a switch called Rate Limiter, and it spaces Big-AGI's own requests about a second apart. Readers usually arrive wanting open-weights models without the hardware to run them (models on your own machine is the other route).

The credit-funded account

Credit is purchased in the console and drawn down per request. Sign-up grants have changed more than once, so read the console rather than a figure quoted elsewhere. Top up before the first answer. Purchased credit, plus whatever limits the console's billing area offers, is what binds. Together states it does not train on your data without explicit opt-in, and offers zero data retention as a setting (privacy).

Where the key comes from

  1. Open api.together.ai/settings/projects/~current/api-keys. The app's older .xyz link still resolves (quickstart).
  2. Create the key.
  3. Copy it before the page changes.

Adding it, and the Rate Limiter switch

Press Ctrl + Shift + M for Models, then More Services if the Setup AI Models wizard opens first. Add -> Together AI, paste into Together AI Key, press Models, and the list appears below the panel (add and manage your keys).

If answers then come back refused for speed - 429 Too Many Requests - turn on Rate Limiter under Advanced. It is Big-AGI's own throttle, and it stops a free-tier account overrunning its allowance.

Direct Connection

Direct Connection can be turned on once the key is in the browser. It starts off.

Note

Direct Connection works only with your API key stored in the browser, and with an AI service that permits direct browser calls (CORS). Where it cannot be used, requests route through the Big-AGI fast edge servers instead - everything still works, within the standard upload size and time limits.

© 2026 Token Fabrics·Built with passion in San Diego