• Playground
  • Models
  • Rankings
    • About Boundless
    • Quickstart
    • Coding Agents & Harnesses
    • Sandbox
    • API Compatibility
    • System One
    • Credits & Billing
    • Pricing
    • Rate Limits
    • Teams
    • FAQ
  • Support
All systems operational
© 2026 BoundlessTermssupport@boundless.network

FAQ

When to use Boundless, when not to, the wire formats, prepaid credits, and what is not served.

Short answers to common questions. The quickstart gets you to a first token, and llms.txt has the same fit guidance written for agents.

When should I use Boundless?#

Use Boundless when you already call an OpenAI or Anthropic chat API and want the same code to run open-weight models. The wire formats are the ones your SDK already sends, so the only change is a base URL and a key.

Use it when you need:

  • the exact cost of each request: every response carries token counts that reproduce the charge against the rate card
  • a hard spending ceiling: credits are prepaid, requests stop at zero, and a key can carry its own budget
  • to point a coding agent or harness (Claude Code, Codex CLI, OpenCode, and others) at cheaper models without changing the harness
  • long context, or tool use and reasoning on open weights

When should I not use Boundless?#

Look elsewhere if:

  • you need provider-native Responses features such as OpenAI-hosted tools or remote conversation storage. /v1/responses serves only the Codex-compatible subset
  • you need embeddings, legacy completions (/v1/completions), fine-tuning, or image or audio generation, none of which are served
  • you need a proprietary model that is not on Models. Most of the lineup is open-weight, with a few models from Anthropic and OpenAI

A capped agent key is mintable instantly for smoke tests; a human account is only needed for real spend.

Which wire formats does the API speak?#

One service speaks three wire formats (OpenAI chat completions, the Codex-compatible Responses subset, and the Anthropic Messages API) over the same models and the same rate card.

  • An OpenAI SDK or Codex custom provider takes the gateway /v1 base.
  • Anthropic SDKs take the gateway root, because they append /v1/messages themselves.
  • Codex uses wire_api = "responses".

Authenticate with Authorization: Bearer, or with x-api-key if that is what your client already sends. Name a model from the catalog. See API Compatibility.

How do prepaid credits work?#

You pay with prepaid credits. Requests debit the shared Team balance exactly. Every charge can be recomputed from the rate card and each response's usage object. A single API key can carry its own budget below the team's. Only Team admins can buy credits or change the card on file.

What happens when credits reach zero?#

Requests stop. There is no overage and no end-of-month invoice. Exhausted credits return 429 budget_exceeded and are not charged past the balance. Waiting does not help; top up instead. The console warns as the balance runs low.

Can I try the API without an account?#

Yes. POST /api/agent-key with no body and no credential, or use the create_agent_key MCP tool. The key is for smoke tests: it has a small cap, expires, and cannot be topped up or extended. GET on the same URL describes the offer without minting.

Keys for real work are created in the console once an account is active.

Can I sign up myself?#

Yes. Sign-up and key creation are self-serve, and a new account is designed to be active the moment you sign up. Adding a card grants $1 of free credits. The model lineup, prices, and base URL may change while the first accounts are onboarded.

Which models does Boundless serve?#

Mostly open-weight models, plus a few proprietary models from Anthropic and OpenAI. Call each one by the API identifier listed on Models.

Each model's context window, capabilities, and live rates are on its page there. Prices are not repeated here, so there is no second copy to go out of date.

How long is the context window?#

Each model publishes its own window on its catalog page. Most served models take about a million tokens; the rest take 262,144. Prompt and completion share that window. Check the live model page for the current number.

What does Boundless not serve?#

These are not served (see API Compatibility):

  • embeddings (/v1/embeddings)
  • legacy completions (/v1/completions)
  • fine-tuning
  • image generation and audio generation
  • Responses features outside the Codex-compatible subset, such as OpenAI-hosted tools or remote conversation storage

Batch inference with 1h and 24h completion windows launches in the coming weeks; the discounted rates are on Pricing.

Can I point a coding agent or harness at Boundless?#

Yes. curl -fsSL https://boundless.md/install | sh signs you in, creates a key, and configures Claude Code, Codex CLI, OpenCode and most other harnesses in one step. Any other client that lets you set a base URL and an API key works too; its wire format decides which base URL you give it. See Coding Agents & Harnesses.