V-APIbeta
  • Console
  • Models
  • Chat
  • Blog
  • Help Center
  • API Docs
Sign InSign Up
  3. Gemini
  5. gemini-3-flash-preview-thinking

gemini-3-flash-preview-thinking

文本对话图像分析文件分析深度思考

Context Length

1M

Max Output

66K

Published

December 18, 2025

Knowledge Cutoff

January 1, 2025

Model / Completion Ratio

0.25 / 6.00

Description

After using the "-thinking" suffix in gemini-3-flash-preview, the system will be forced to think and display the model's thought process.

Available Groups

Different token groups have different prices. Unit: Million tokens (M)

Token GroupDescriptionGroup RatioInput PriceOutput Price
geminiGemini企业渠道（官方T3号池）1.5x$0.75/M$4.5/M
gemini_cliGemini性价比（来源cli）0.5x$0.25/M$1.5/M

Supported Endpoints

/v1/chat/completionsPOST
/v1/messagesPOST
/v1beta/models/gemini-3-flash-preview-thinking:generateContentPOST

"V-API is a high-quality large language model API service platform, dedicated to providing stable, fast, and secure API testing solutions for individuals, developers, and enterprises."

Help
Help CenterAPI DocsTicket Service
Team
Business ContactBlogPartner
协议
免责声明隐私政策服务条款
Social
© 2023-2026 V-API, All rights reserved This site's API is for testing and experience purposes. Please comply with your local laws and regulations. Do not use for illegal purposes. This site assumes no legal liability.