Create a chat completion

POST

/chat/completions

curl --request POST \
  --url https://api.saygm.com/v1/chat/completions \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{ "model": "gpt-5.5", "messages": [ { "role": "system", "content": "example" } ], "stream": false, "temperature": 1, "max_tokens": 1 }'

OpenAI-compatible Chat Completions. Set stream: true for an incremental Server-Sent Events stream of chat.completion.chunk objects terminated by data: [DONE]; otherwise a single chat.completion object is returned.

Authorizations

• bearerAuth

Request Body^required

Media typeapplication/json

object

model

required

Bare model id (no provider prefix); the gateway resolves the provider.

string

messages

required

Array<object>

>= 1 items

object

role

required

string

Allowed values: system user assistant tool

content

required

string

stream

Stream the response as Server-Sent Events.

boolean

temperature

number

max_tokens

integer

key

additional properties

any

Responses

200

When stream is false, a single chat.completion JSON object. When stream is true, a Server-Sent Events stream (text/event-stream) of chat.completion.chunk objects terminated by data: [DONE].

object

id

string

object

string

Allowed value: chat.completion

created

integer

model

string

choices

Array<object>

object

index

integer

message

object

role

required

string

Allowed values: system user assistant tool

content

required

string

finish_reason

string

usage

object

prompt_tokens

integer

completion_tokens

integer

total_tokens

integer

cost

USD settled against the account for this request — the gateway’s settlement figure, not a catalog estimate. Rendered with exactly nine fractional digits.

number

cost_nano_usd

The same settled amount in nano-USD (10^-9 USD), gm’s exact integer accounting unit.

integer

Example

{
  "object": "chat.completion",
  "choices": [
    {
      "message": {
        "role": "system"
      }
    }
  ]
}

SSE stream of chat.completion.chunk objects, one per data: line, terminated by data: [DONE].

string

400

Invalid request (e.g. unknown model).

Media typeapplication/json

object

error

object

message

string

type

string

code

string | null

Example^generated

{
  "error": {
    "message": "example",
    "type": "example",
    "code": "example"
  }
}

401

Missing or invalid gm API key.

Media typeapplication/json

object

error

object

message

string

type

string

code

string | null

Example^generated

{
  "error": {
    "message": "example",
    "type": "example",
    "code": "example"
  }
}

402

Insufficient prepaid balance.

Media typeapplication/json

object

error

object

message

string

type

string

code

string | null

Example^generated

{
  "error": {
    "message": "example",
    "type": "example",
    "code": "example"
  }
}