Source: https://01router.com/docs/models/gemini-3.8-flash/chat
Base URL: https://01router.com/api/v1

# Gemini 3.8 Flash · Chat completions

Chat models (GPT, Claude, Gemini, Grok, DeepSeek…) use the OpenAI Chat Completions format, so the official OpenAI SDKs work with base URL https://01router.com/api/v1. Set "stream": true for server-sent events. Credits for the prompt plus max_tokens (4,096 when unset) are held first, then settled to the real token usage once the reply ends; the same request id appears in the X-Gateway-Request-Id header and the task log.

`POST /api/v1/chat/completions`

Bearer API key

## Input parameters

| Field | Type | Required | Description | Constraints |
| --- | --- | --- | --- | --- |
| model | string | yes | Model slug from the catalog. Project keys may use a project-specific model slug. | options: [   "gemini-3.8-flash" ] |
| messages | object[] | yes | Non-empty message array with role and content; content may contain supported multimodal parts. |  |
| provider | string | no | Optional provider slug. Omit for automatic routing; project settings may pin the provider. |  |
| project | string | no | Optional project ID. Use its published model catalog and routing configuration. |  |
| stream | boolean | no | Default false. true returns text/event-stream chunks and ends with data: [DONE]. | default: false |
| max_tokens | integer | no | Positive integer. Default 4096. Do not combine with max_completion_tokens. | default: 4096; min: 1 |
| max_completion_tokens | integer | no | Positive integer. Default 4096. Do not combine with max_completion_tokens. |  |
| temperature | number | no | OpenAI-compatible sampling, tools and structured output fields depend on the selected model and provider. Only n = 1 is supported. |  |
| tools | object[] | no | OpenAI-compatible sampling, tools and structured output fields depend on the selected model and provider. Only n = 1 is supported. |  |
| response_format | object | no | OpenAI-compatible sampling, tools and structured output fields depend on the selected model and provider. Only n = 1 is supported. |  |

## Response fields

| Field | Type | Required | Description | Constraints |
| --- | --- | --- | --- | --- |
| id | string | — | Generation ID returned by the create request. |  |
| object | string | — | chat.completion / chat.completion.chunk |  |
| model | string | — | Model slug from the catalog. Project keys may use a project-specific model slug. |  |
| choices | array | — | Completion choices. Non-streaming replies use choices[].message; streaming chunks use choices[].delta. |  |
| usage | object | — | prompt_tokens, completion_tokens and total_tokens. Streaming includes usage when supplied by the provider. |  |

## cURL

```bash
curl 'https://01router.com/api/v1/chat/completions' \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "gemini-3.8-flash",
  "messages": [
    {
      "role": "user",
      "content": "Hello!"
    }
  ],
  "max_tokens": 512,
  "stream": false
}'
```

## 200 · application/json

```json
{
  "id": "chat_example",
  "object": "chat.completion",
  "model": "gemini-3.8-flash",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Hello!"
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 8,
    "completion_tokens": 2,
    "total_tokens": 10
  }
}
```

## SSE · stream: true

```text
data: {"id":"chat_example","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"content":"Hello!"},"finish_reason":null}]}

data: [DONE]
```

## Usage notes

- OpenAI-compatible sampling, tools and structured output fields depend on the selected model and provider. Only n = 1 is supported.