gemini-2.5-flash Reasoning

POST

/v1/chat/completions

You can use the reasoning_effort and thinking fields to control reasoning length

Request

Authorization

API Key

Add parameter in header

Authorization

Example:

Authorization: ********************

Body Params application/json

Examples

Use reasoning_effort to control reasoning levelUse budget_tokens to control maximum reasoning token length

{
    "model": "gemini-2.5-flash",
    "messages": [
        {
            "content": [
                {
                    "text": "What is the sun",
                    "type": "text"
                }
            ],
            "role": "user"
        }
    ],
    "reasoning_effort": "minimal"
}

Request Code Samples

Shell

JavaScript

Java

Swift

PHP

Python

HTTP

Objective-C

Ruby

OCaml

Dart

cURL

Use reasoning_effort to control reasoning levelUse budget_tokens to control maximum reasoning token length

curl --location 'https://api-xmodel.nexconn.ai/v1/chat/completions' \
--header 'Authorization: <api-key>' \
--header 'Content-Type: application/json' \
--data '{
    "model": "gemini-2.5-flash",
    "messages": [
        {
            "content": [
                {
                    "text": "What is the sun",
                    "type": "text"
                }
            ],
            "role": "user"
        }
    ],
    "reasoning_effort": "minimal"
}'

Responses

🟢200Success

application/json

Bodyapplication/json

Example

{}

Modified at 2026-04-24 10:39:56

openai/gpt-5.2-codex

claude-4.5-sonnet reasoning