← all releases
DeepSeek AI·Jun 27, 2026·v1.0·open source
DeepSeek V4 Pro DSpark
DeepSeek V4 Pro DSpark is an advanced mixture-of-experts (MoE) language model optimized for high-efficiency, million-token context intelligence. It features a specialized speculative decoding module to accelerate inference performance while maintaining state-of-the-art reasoning and agentic capabilities.
LLMMoELong-ContextOpen-Source
overview
DeepSeek V4 Pro DSpark is an advanced mixture-of-experts (MoE) language model optimized for high-efficiency, million-token context intelligence. It features a specialized speculative decoding module to accelerate inference performance while maintaining state-of-the-art reasoning and agentic capabilities.
key features
- 011.6 Trillion parameter MoE architecture (49B activated)
- 02Support for 1 million token context length
- 03Hybrid Attention Architecture (CSA and HCA) for efficiency
- 04Speculative decoding module integration
- 05Supports multiple reasoning effort modes (Non-think, High, Max)
- 06Optimized using the Muon optimizer for faster convergence
related productsbrowse all →
Protenixopen sourceProtenix is an open-source, trainable PyTorch implementation of AlphaFold 3, designed for high-accuracy biomolecular structure prediction. It aims to advance accessible and extensible research tools for the computational biology community. ([github.com](https://github.com/bytedance/Protenix?utm_source=openai))Z-Imageopen sourceZ-Image is an efficient image generation foundation model developed by Tongyi-MAI, designed to produce high-quality, diverse, and stylistically versatile images. It serves as a robust base for creators, researchers, and developers seeking advanced image generation capabilities.