> ## Documentation Index
> Fetch the complete documentation index at: https://docs.loom.aayanmishra.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Endpoints

> OpenAI-compatible endpoints implemented by Loom AI API.

# Endpoints

Loom AI API mirrors the common OpenAI endpoints with the following notable routes:

* `/completions` — legacy completions endpoint (sync/async)
* `/responses` — modern response-completions hybrid
* `/chat/completions` — conversational/chat completions
* `/embeddings` — embeddings
* `/moderations` — content moderation

### Example: chat completions (curl)

```bash theme={null}
curl https://api.loom.aayanmishra.com/api/v1/chat/completions \
  -H "Authorization: Bearer $LOOM_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-4o-mini","messages":[{"role":"user","content":"Summarize the following..."}]}'
```

### Example: embeddings

```bash theme={null}
curl https://api.loom.aayanmishra.com/api/v1/embeddings \
  -H "Authorization: Bearer $LOOM_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"text-embedding-3-small","input":"Hello world"}'
```

### HTTP semantics

All endpoints follow standard OpenAI response shapes where possible. Rate limiting and errors follow OpenAI-like codes but include Loom-specific fields in the `error` object (see errors page).