DeepSeek🇨🇳
A Chinese AI research lab known for developing high-performance, cost-effective, open-weights LLMs like DeepSeek-R1 and V3, which rival top U.S. models in reasoning, coding, and math.
Timeline
mHC: Manifold-Constrained Hyper-Connections
DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
Insights into DeepSeek-V3: Scaling Challenges and Reflections on Hardware for AI Architectures
DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models
DeepSeekMath-V2: Towards Self-Verifiable Mathematical Reasoning
DeepSeek-OCR: Contexts Optical Compression
Inference-Time Scaling for Generalist Reward Modeling
DeepSeek-Prover-V2: Advancing Formal Mathematical Reasoning via Reinforcement Learning for Subgoal Decomposition
CodeI/O: Condensing Reasoning Patterns via Code Input-Output Prediction
JanusFlow: Harmonizing Autoregression and Rectified Flow for Unified Multimodal Understanding and Generation
Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention
DeepSeek-V3 Technical Report
Janus-Pro: Unified Multimodal Understanding and Generation with Data and Model Scaling
DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understanding
Janus: Decoupling Visual Encoding for Unified Multimodal Understanding and Generation
Fire-Flyer AI-HPC: A Cost-Effective Software-Hardware Co-Design for Deep Learning
Auxiliary-Loss-Free Load Balancing Strategy for Mixture-of-Experts
DeepSeek-Prover-V1.5: Harnessing Proof Assistant Feedback for Reinforcement Learning and Monte-Carlo Tree Search
Let the Expert Stick to His Last: Expert-Specialized Fine-Tuning for Sparse Architectural Large Language Models
FAQ
What does DeepSeek timeline include?
The DeepSeek timeline includes public content currently linked to this entity on Attendemia, ordered by recent publishing and activity signals. It is designed to show the most relevant updates first, so readers can quickly understand current work without manually scanning multiple external sources.
How are items linked to this entity?
Items are linked through source ingestion, entity slug matching, and curation checks that validate association quality. Attendemia combines crawl signals with tag mapping before showing content on this page, which helps reduce noisy matches and keeps the entity timeline focused on relevant research outputs.
Where can I explore related topics?
Use the related topics block on this entity page to jump into adjacent topic hubs and broaden your discovery graph. These topic links are generated from shared content overlap, so they surface nearby areas that frequently co-occur with this entity’s publications and references.
Can I follow this entity for updates?
Yes. Use the follow button near the page title to track future timeline additions for this entity. Following creates a lightweight monitoring workflow, so you can revisit important updates quickly and reduce repeated manual checks across multiple websites, feeds, and publication channels.