DeepSeek-V4 Preview: Entering the Era of Million-Context Accessibility
Today, the preview version of our brand-new model series DeepSeek-V4 is officially launched and open-sourced simultaneously.
DeepSeek-V4 features million-word ultra-long context, leading both domestically and in the open-source field in Agent capabilities, world knowledge, and reasoning performance. The model is divided into two versions based on size:
Starting today, you can chat with the latest DeepSeek-V4 on the official website to explore the new experience of 1M ultra-long context memory. API services have been updated simultaneously, and you can call them by modifying model_name to deepseek-v4-pro or deepseek-v4-flash.
DeepSeek-V4-Pro: Performance Comparable to Top Closed-Source Models
Significantly improved Agent capabilities: Compared to previous generation models, DeepSeek-V4-Pro’s Agent capabilities are remarkably enhanced. In Agentic Coding evaluations, V4-Pro has reached the best level among current open-source models and performs excellently in other Agent-related evaluations. Currently, DeepSeek-V4 has become the Agentic Coding model used by internal employees. According to evaluation feedback, the user experience is better than Sonnet 4.5, and delivery quality is close to Opus 4.6 non-thinking mode, though there is still a certain gap with Opus 4.6 thinking mode.
Rich world knowledge: In world knowledge assessments, DeepSeek-V4-Pro significantly leads other open-source models, only slightly inferior to the top closed-source model Gemini-Pro-3.1.
World-class reasoning performance: In evaluations of mathematics, STEM, and competition-level code, DeepSeek-V4-Pro surpasses all currently publicly evaluated open-source models, achieving excellent results comparable to the world’s top closed-source models.
DeepSeek-V4-Flash: A Faster and More Efficient Cost-Effective Option
Compared to DeepSeek-V4-Pro, DeepSeek-V4-Flash is slightly inferior in world knowledge reserve but demonstrates comparable reasoning capabilities. Due to smaller model parameters and activation, V4-Flash can provide faster and more cost-effective API services.
In Agent evaluations, DeepSeek-V4-Flash is on par with DeepSeek-V4-Pro in simple tasks but still has gaps in high-difficulty tasks.
Structural Innovation and Ultra-High Context Efficiency
DeepSeek-V4 has pioneered a brand-new attention mechanism that compresses in the token dimension, combined with DSA sparse attention (DeepSeek Sparse Attention), achieving world-leading long context capabilities and significantly reducing computing and memory requirements compared to traditional methods. From now on, 1M (one million) context will be the standard for all official DeepSeek services.
Special Optimization for Agent Capabilities
DeepSeek-V4 has been adapted and optimized for mainstream Agent products such as Claude Code, OpenClaw, OpenCode, CodeBuddy, etc., with improved performance in code tasks, document generation tasks, and more.
API Access
Currently, DeepSeek API has simultaneously launched V4-Pro and V4-Flash, supporting OpenAI ChatCompletions interface and Anthropic interface. When accessing the new models, base_url remains unchanged, and the model parameter needs to be changed to deepseek-v4-pro or deepseek-v4-flash.
V4-Pro and V4-Flash have a maximum context length of 1M, both supporting non-thinking mode and thinking mode, where thinking mode supports the reasoning_effort parameter to set thinking intensity (high/max). For complex Agent scenarios, it is recommended to use thinking mode and set the intensity to max.
Please note: The two old API interface model names deepseek-chat and deepseek-reasoner will be discontinued in three months (2026-07-24). At the current stage, these two model names point to the non-thinking mode and thinking mode of deepseek-v4-flash respectively.
Open-Source Weights and Local Deployment
DeepSeek-V4 model open-source links:
- Hugging Face: https://huggingface.co/collections/deepseek-ai/deepseek-v4
- ModelScope: https://modelscope.cn/collections/deepseek-ai/DeepSeek-V4
DeepSeek-V4 technical report: https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro/blob/main/DeepSeek_V4.pdf