Boundless inference for open-weight models
An inference platform for open-weight models, sold on prepaid credits. One service speaks three wire formats — OpenAI chat completions, the Codex-compatible Responses subset, and the Anthropic Messages API — over the same models on the same rate card.
Use the API
Boundless serves a curated model lineup through an OpenAI-compatible chat completions API and the Anthropic Messages API. Existing clients and coding agents can connect with familiar request formats. The platform supports streaming responses, tool use, and exact input and output token counts.
Agents can request a short-lived trial credential without an account or browser. Send an empty POST request to https://inference.boundless.network/api/agent-key, then use the returned key with the documented inference endpoints.
Predictable usage and billing
Access is sold on prepaid credits. Requests stop when the shared Team balance reaches zero instead of creating an unbounded invoice. Published per-token rates and exact usage make every request cost reproducible. API keys can have individual spend budgets, while Team accounts share credits, usage, and billing across members.
Start here
Read the quickstart, the FAQ, check API compatibility, inspect the current model pricing, browse the models we serve, or fetch llms.txt for the complete agent-oriented index. The OpenAPI document describes the public inference surface programmatically.
Boundless Networks, Inc. operates this service. For customer support, email support@boundless.network or use the support page.