> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.astropods.com/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.astropods.com/_mcp/server.

# Errors and limits

The gateway returns standard OpenAI-style error responses. Common cases:

| Status | Meaning                                  | What to do                                                           |
| ------ | ---------------------------------------- | -------------------------------------------------------------------- |
| `401`  | Missing/invalid `ASTRO_GATEWAY_API_KEY`. | Redeploy so the key is re-injected; don't hardcode it.               |
| `404`  | Wrong path.                              | Ensure your base URL ends in `/v1`.                                  |
| `429`  | Rate or usage limit reached.             | Back off and retry; the OpenAI SDKs retry `429`/`5xx` automatically. |

## What the gateway doesn't cover

* **Provider-native SDKs and prefixes (except Anthropic).** Call models through the OpenAI-compatible API by their gateway id (for example `claude-sonnet-4-6`), not with a provider prefix such as `anthropic/`. The one exception is the native Anthropic Messages API, served on the `/anthropic` passthrough for Claude Code / Claude Agent SDK agents. See [Anthropic API](/ai-gateway/anthropic).
* **Anthropic server-side tools.** `web_search`, `web_fetch`, code execution, and computer use aren't available on the Claude models. Client-defined tools work normally, and the `gpt-*` models have their own [web search](/ai-gateway/web-search).
* **Image generation.** No image-generation models are offered. Image *input* is supported: see [Vision](/ai-gateway/vision).
* **Bring-your-own-model.** Adding new models or fine-tunes to the gateway isn't self-serve; [contact support](/support).

## Next steps

* [Usage limits](/usage-limits): cap what an account can spend before it hits `429`
* [Usage](/usage): see what your account has spent on model calls this period