Skip to main content
Tama provides on-demand Linux machines (optionally with GPUs) that snapshot on stop and can be forked, controlled through its CLI and SDKs. The following table shows what CostGraph runs on this provider.

​
Connect

Connect Tama on Integrations, and CostGraph starts sandboxes in your Tama account. You don’t run anything yourself.
  1. In CostGraph, go to Integrations > Sandbox > Tama.
  2. Enter these fields:
    • API token (TAMA_TOKEN, required): A Tama API token, created with tama tokens create.
    • API URL (TAMA_API_URL, optional): A different Tama API address, if you do not use the default.
  3. Click Connect.

​
Your own dispatcher

To keep the key out of CostGraph or reach a private network, run your own dispatcher instead. Set provider to tama and the preceding fields as environment variables, as described in Run your own dispatcher.

​
Key and permissions

TAMA_TOKEN is a Tama API token, created with “tama tokens create “—name ci“ (or tama.tokens.create in the SDK); the secret is printed once. The provider doesn’t document key scopes or roles, so a key has full access to the account. Sources: tama.computer: llms.txt, tama.computer: docs.

​
Limits

  • A machine has no idle timeout unless requested (min 60 s); CostGraph turns auto-stop off. (source)
  • 1 vCPU and 1024 MiB are floors; no maximum CPU/memory, lifetime, regions or architecture are documented. /workspace is a 50 GiB durable disk. (source)