deepseek-v3.2

DeepSeek-V3.2 is a major large language model release scheduled for December 1, 2025. It comes with both a formal version optimized for everyday applications and a Specialized version enhanced for long-form reasoning. This model achieves a key breakthrough in reasoning capabilities, reaching GPT-5-level performance, and for the first time integrates reasoning abilities with tool-call capabilities. With open-source availability and rich features, it adapts to diverse scenarios. Meanwhile, the Specialized version has achieved outstanding results in top international competitions, pushing the performance of open-source models to a new level.

Basic Information

• Released simultaneously with DeepSeek-V3.2-Specialized on December 1, 2025; the former is the official version optimized for everyday applications, while the latter is the long-form reasoning-enhanced edition.
• The official website, app, and API services have all been fully updated to version V3.2. Both models are open-sourced on the HuggingFace and ModelScope platforms.
• The Specialized version is available via a temporary API until December 15, 2025, with a default maximum output length of 128K and supports only thought-based conversations.

Core Features

• Reasoning capability reaches GPT-5 level; its performance on mainstream reasoning benchmarks is slightly below Gemini-3.0-Pro, yet its output length is significantly shorter than Kimi-K2-Thinking.
• The first version to integrate both reasoning and tool calls, supporting dual-mode tool usage—both reasoning and non-reasoning—and adding support for Claude Code.
• The Specialized version won gold medals in four top international competitions, including IMO 2025 and CMO 2025; its ICPC and IOI results ranked second and tenth, respectively, among human competitors.

Technical Highlights

• Adopts the DSA sparse attention mechanism, validated by the V3.2-Exp experimental version, showing no performance degradation compared to V3.1-Terminus in specific scenarios.
• Constructed over 1,800 environments and more than 85,000 complex instructions for reinforcement learning tasks, enabling large-scale agent training data synthesis.
• The Specialized version integrates DeepSeek-Math-V2’s theorem-proving capabilities, providing rigorous mathematical proof and logical verification abilities.

Market Impact

• The agent evaluation reaches the highest level among open-source models, significantly narrowing the gap between open-source and closed-source models in tool-call capabilities.
• The Specialized version achieves breakthrough performance in top international competitions, driving the extreme enhancement of reasoning capabilities in open-source models.

Application Scenarios

• The V3.2 version is suitable for question-answering scenarios and general agent tasks, balancing reasoning ability and efficiency while reducing computational costs.
• The Specialized version is ideal for high-end research scenarios such as mathematical proofs and logical verification; it is not yet optimized for everyday conversations or writing tasks.
• Supports multi-scenario tool calls; in thought mode, multiple rounds of reasoning combined with tool calls can produce more accurate responses.

Huggingface open-source link: https://huggingface.co/deepseek-ai/DeepSeek-V3.2

ModelScope (Mota Community) open-source link: https://modelscope.cn/models/deepseek-ai/DeepSeek-V3.2

If you’d like to experience the ultimate performance of the Specialized version, please visit deepseek-v3.2-speciale