API reference · Chat Completions

Chat Completions.

The OpenAI Chat Completions dialect on the lx1 catalog. If your client speaks to OpenAI, it speaks to Layer X1 — swap the base URL and the key, keep everything else.

Endpoint

POST/v1/chat/completions

Authenticate with Authorization: Bearer lx1_.... Any model in the catalog is valid in model.

Request parameters

Parameter	Type	Notes
`model`	string · required	Any catalog id, e.g. `lx1-gpt-oss-120b`.
`messages`	array · required	Chat history. Content can be a string or content parts; image parts (`image_url`, data URLs included) are accepted on vision models.
`max_tokens`	integer	Output ceiling. On reasoning models, hidden reasoning counts against it — budget generously.
`temperature`	number	Sampling temperature.
`top_p`	number	Nucleus sampling.
`stop`	string \| string[]	Stop sequences.
`stream`	boolean	SSE chunks when `true`. See Streaming.
`stream_options`	object	`{ "include_usage": true }` appends a final usage chunk.
`tools`	array	Function tools. See Tool calling.
`tool_choice`	string \| object	`auto`, `none`, `required`, or a named function.
`response_format`	object	JSON mode / `json_schema`. See Structured output.

Example

curl https://api.layerx1.in/v1/chat/completions \
  -H "authorization: Bearer $LAYERX1_API_KEY" \
  -H "content-type: application/json" \
  -d '{
    "model": "lx1-gpt-oss-120b",
    "messages": [
      { "role": "system", "content": "You are concise." },
      { "role": "user", "content": "Say hi." }
    ]
  }'

Create a chat completion

{
  "id": "chatcmpl_...",
  "object": "chat.completion",
  "model": "lx1-gpt-oss-120b",
  "choices": [{
    "index": 0,
    "message": { "role": "assistant", "content": "Hi." },
    "finish_reason": "stop"
  }],
  "usage": { "prompt_tokens": 14, "completion_tokens": 3, "total_tokens": 17 }
}

Response (shape)

Response fields

Field	Notes
`choices[].message`	The assistant turn — `content`, and `tool_calls` when the model calls a tool.
`choices[].finish_reason`	`stop`, `length` (hit `max_tokens`), or `tool_calls`.
`usage`	Prompt, completion, and total token counts for the request.

Streaming

With stream: true the response is Server-Sent Events: incremental chat.completion.chunk objects, terminated by data: [DONE].

curl -N https://api.layerx1.in/v1/chat/completions \
  -H "authorization: Bearer $LAYERX1_API_KEY" \
  -H "content-type: application/json" \
  -d '{
    "model": "lx1-gpt-oss-120b",
    "stream": true,
    "stream_options": { "include_usage": true },
    "messages": [{ "role": "user", "content": "Count to five." }]
  }'

Stream tokens as they generate

Errors

Failures return standard status codes with OpenAI-shaped bodies — match on the status, retry on 429/5xx with backoff. Full reference: Errors.

Note

Requesting vision input on a text-only model is a 400; check capability flags on the model catalog first — only lx1-sonnet-4.6 reads images.