Skip to main content
POST
Create chat completion (pod OpenAI adapter)

Drop-in with the OpenAI SDK

The pod adapter is OpenAI-compatible. Point any OpenAI SDK at your pod’s /v1 base URL, use the pod integration key (sk-sglpod-int-...) as the API key, and call the model agent-pod. Each request runs one agent turn on your pod.
Non-streaming only. Setting stream: true returns 400 stream_unsupported. The response usage field is always null in v1.
Base URL: https://compute.x402layer.cc/pods/{id}/v1

Quickstart (cURL)

Headers

Authorization
string
required

Pod integration key as a bearer token.

Example:

"Bearer sk-sglpod-int-..."

Path Parameters

id
string<uuid>
required

Body

application/json
messages
object[]
required
Minimum array length: 1
Example:
model
string

Ignored by the adapter; use agent-pod.

Example:

"agent-pod"

stream
boolean
default:false

Must be false or omitted. true returns 400 stream_unsupported.

Response

Chat completion

id
string
Example:

"chatcmpl_sgl_..."

object
string
Example:

"chat.completion"

created
integer
model
string
Example:

"agent-pod"

choices
object[]
usage
null

Always null in v1.