Chat Completion

Create a chat completion with a GreenPT model. We recommend gemma4: long-context, multimodal, and hosted on sustainable EU infrastructure. See Models for the full list.

POST/v1/chat/completions

Authorization

bearerAuth

AuthorizationBearer <token>

In: header

Request Body

application/json

TypeScript Definitions

Use the request body type in TypeScript.

model*string

ID of the model to use.

Default"gemma4"

messages*array<>

A list of messages comprising the conversation so far.

stream?boolean

If set, partial message deltas will be sent as server-sent events.

Defaultfalse

temperature?number

Sampling temperature. Higher = more random.

Default1

Formatfloat

Range0 <= value <= 2

top_p?number

Nucleus sampling probability mass.

Formatfloat

Range0 <= value <= 1

max_tokens?integer

Maximum number of tokens to generate.

Range1 <= value

n?integer

Number of completions to generate for each prompt.

Default1

Range1 <= value

stop?|array<string>

Stop sequence(s) where generation will halt.

presence_penalty?number

Default0

Formatfloat

Range-2 <= value <= 2

frequency_penalty?number

Default0

Formatfloat

Range-2 <= value <= 2

user?string

A unique identifier representing your end-user.

reasoning_effort?string

Controls the model's thinking budget. When omitted, thinking is enabled by default. Pass "none" to disable thinking entirely. The values minimal / low / medium / high are accepted for OpenAI-spec compatibility and all currently map to thinking-enabled on GreenPT-hosted models.

Value in"none" | "minimal" | "low" | "medium" | "high"

Response Body

`application/json`

curl -X POST "https://example.com/v1/chat/completions" \  -H "Content-Type: application/json" \  -d '{    "model": "gemma4",    "messages": [      {        "role": "user",        "content": "Hello, how are you?"      }    ]  }'

{
  "id": "string",
  "object": "chat.completion",
  "created": 0,
  "model": "string",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "string",
        "reasoning_content": "string",
        "tool_calls": [
          {}
        ]
      },
      "logprobs": {},
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 0,
    "completion_tokens": 0,
    "total_tokens": 0
  }
}

{
  "error": {
    "code": 0,
    "message": "string",
    "type": "string",
    "param": "string"
  },
  "requestId": "string"
}

{
  "error": {
    "code": 0,
    "message": "string",
    "type": "string",
    "param": "string"
  },
  "requestId": "string"
}

Notes

Recommended model: gemma4. Other models (e.g. green-l, green-r) are available, see Models.
Streaming: set stream: true and consume the response as Server-Sent Events. See Chat Completion (Streaming).
Authorization: pass your API key in the Authorization header as Bearer YOUR_API_KEY.
Reasoning effort: reasoning_effort is supported on GreenPT-hosted models. Omit it to keep thinking enabled (the default), or pass "none" to disable thinking entirely. OpenAI-spec values (minimal / low / medium / high) are accepted for compatibility and all currently map to thinking-enabled.

Authorization

Request Body

Response Body

200application/json

401application/json

429application/json

Notes

`application/json`

`application/json`

`application/json`