ϟDomestic Models MiniMax Releases M3.1 Flash: 3x Performance Boost, 70% Cost Reduction Learn MoreϟDomestic Models Kimi K3 Released and API Opened Read More →✦New Model Launch Wan3.0 Released View Now →✦New Model Launch DeepSeek V4.1 Flash: Native Multimodal, Outperforms V4 Pro with Lower API Pricing View Details✦New Model Online Seedance 2.5 Release Details View Now →ϟDomestic Models MiniMax Releases M3.1 Flash: 3x Performance Boost, 70% Cost Reduction Learn More
AGTCloud
Contact us Careers
Back to updates
DeepSeek

DeepSeek V4 Pro Preview Release: Million-Token Context and Enhanced Agent Capabilities

DeepSeek V4 Pro is designed for complex reasoning, mathematics, coding, and agent tasks, with the official API offering a 1M context window and both thinking and non-thinking modes.

DeepSeek V4 Pro release visual

DeepSeek V4 Pro and V4 Flash are now available via the official API. The Pro version focuses on enhancing capabilities in mathematics, STEM, coding, and agents.

The official documentation requires setting the model parameter to deepseek-v4-pro; pricing, rate limits, and migration timelines for older models should be verified against the latest DeepSeek documentation.

KEY HIGHLIGHTS

Key highlights

  • 1M token context window

  • Supports thinking and non-thinking modes

  • OpenAI and Anthropic compatible interfaces

RELATED UPDATES

Related updates