# LLM API guides

Documentation for using LLM API's agent storefront: what it sells, how to call it, what it costs, and how to recover from errors. The storefront itself: https://agents.llmapi.ai/zcj/57ug1d1rg9rj/llms.txt.

- [Create text embeddings with LLM API's embeddings API](https://agents.llmapi.ai/guides/embeddings.md): One endpoint serves Text Embedding 3 Small, Text Embedding 3 Large and Jina Embeddings v4, billed per input token from prepaid credit.
- [Chat Completions, Anthropic Messages or Responses API on LLM API](https://agents.llmapi.ai/guides/request-formats.md): Every LLM API chat model answers on Chat Completions, Anthropic Messages and the Responses API. Keep your request format and change only the model id.
- [Call Claude, GPT, Gemini and DeepSeek from one API with LLM API](https://agents.llmapi.ai/guides/quickstart.md): Call Claude, GPT, Gemini and DeepSeek models through one OpenAI-compatible API: one free credential, one prepaid credit balance, billed by exact token usage.
- [How much does LLM API cost per token?](https://agents.llmapi.ai/guides/pricing.md): Every LLM API model is billed by exact token usage at 5% below list price, drawn from one prepaid credit balance. Count tokens first to estimate input.
- [LLM API models: model ids, context sizes and per-token prices](https://agents.llmapi.ai/guides/models.md): LLM API's chat and embedding models with model ids, context sizes and per-token prices, grouped by price. Switch models by changing one field.