DeepSeek-V3.2 is a model that harmonizes high computational efficiency with superior reasoning and agent performance. Its approach is built upon three key technical breakthroughs: DeepSeek Sparse Attention (DSA), an efficient attention mechanism that substantially reduces computational complexity while preserving model performance, specifically optimized for long-context scenarios; a Scalable Reinforcement Learning Framework, which enables performance comparable to GPT-5 and reasoning proficiency on par with Gemini-3.0-Pro in its high-compute variant; and a Large-Scale Agentic Task Synthesis Pipeline to integrate reasoning into tool-use scenarios, improving compliance and generalization in complex interactive environments. The model has achieved gold-medal performance in the 2025 International Mathematical Olympiad (IMO) and International Olympiad in Informatics (IOI)...

DeepSeek-V3.2-Exp is an experimental version of DeepSeek model, built on V3.1-Terminus. It debuts DeepSeek Sparse Attention(DSA) for faster, more efficient training & inference on long context....

DeepSeek-V3.1-Terminus is an updated version built on V3.1’s strengths while addressing key user feedback. It improves in language consistency, reducing instances of mixed Chinese-English text and occasional abnormal characters. And also upgrades in stronger Code Agent & Search Agent performance....

DeepSeek-V3.1 is a hybrid model that supports both thinking mode and non-thinking mode. Through post-training optimization, the model's performance in tool usage and agent tasks has significantly improved. DeepSeek-V3.1-Think achieves comparable answer quality to DeepSeek-R1-0528, while responding more quickly....

DeepSeek-V3-0324 demonstrates notable improvements over its predecessor, DeepSeek-V3, in several key aspects, including major boost in reasoning performance, stronger front-end development skills and smarter tool-use capabilities....

DeepSeek-R1-0528 is an upgraded model shows significant improvements in handling complex reasoning tasks，also offers a reduced hallucination rate, enhanced support for function calling, and better experience for vibe coding. It achieves performance comparable to O3 and Gemini 2.5 Pro....

DeepSeek-V4.1-Flash is DeepSeek's latest multimodal MoE model with 552B backbone parameters. Built on a Causal Encoder-Decoder architecture, it activates only ~8B parameters during prefill and ~16B during decode, delivering flagship-level intelligence at significantly lower inference cost. The model natively understands images, supports a 1M-token context window. It outperforms DeepSeek-V4-Pro across coding, agentic and automation benchmarks, while reducing the KV cache footprint by roughly 4x versus its predecessor. ...

DeepSeek-V4-Flash-Vision-Exp is the first experimental multimodal model in the DeepSeek-V4 family. It builds on the DeepSeek-V4-Flash architecture by incorporating visual modules and undergoing continued training to unlock visual understanding capabilities. It matches the official DeepSeek-V4-Flash on text-only capabilities (agents, reasoning, world knowledge), while achieving substantial improvements on multimodal agent benchmarks, bringing performance close to Opus-4.8. It shares the same model structure as DeepSeek-V4-Flash-DSpark, i.e. it comes with a speculative decoding module attached. Images are tokenized for billing at up to 384 tokens each....

DeepSeek-V4-Pro-0813 is the official release of DeepSeek-V4-Pro, superseding the preview version, with greatly enhanced agentic capabilities and performance improvements that are especially pronounced in production environments. It is built on the DeepSeek-V4-Pro (Preview) model structure, with a DSpark speculative decoding module attached....

DeepSeek-V4-Flash-0731 is the official release of DeepSeek-V4-Flash, superseding the preview version, with substantially enhanced agentic capabilities. It has the same model structure as DeepSeek-V4-Flash-DSpark, i.e. it comes with a speculative decoding module attached....

DeepSeek-V4-Pro is DeepSeek's flagship open-source MoE model with 1.6T total parameters and 49B activated, purpose-built for frontier-level reasoning, coding, and agentic tasks. Supporting a 1M-token context window and three reasoning effort modes up to Think Max, it achieves top-tier performance on coding benchmarks such as LiveCodeBench and Codeforces — rivaling leading closed-source models — and is released under the MIT License....

DeepSeek-V4-Flash is DeepSeek's latest open-source MoE model featuring 284B total parameters with only 13B activated during inference, delivering high-speed generation without sacrificing capability. With native support for a 1M-token context window and three switchable reasoning modes — Non-Think, Think High, and Think Max — it offers flexible intelligence scaling from everyday tasks to complex reasoning, all under the MIT License....