PeopleofSuperintelligence

Researcher

Yaofeng Sun

Yaofeng Sun is a member of the DeepSeek-AI team in Hangzhou, China, and an author of its technical reports on the DeepSeek language models. DeepSeek LLM appeared in January 2024, followed by DeepSeek-VL in March 2024, DeepSeek-V2 in May 2024 and DeepSeek-V3 in December 2024. In September 2025, Nature published the DeepSeek-R1 paper with him among its authors, and in October 2025 he co-authored DeepSeek-OCR with Haoran Wei and Yukun Li. DeepSeek-V3.2 followed in December 2025 and the DeepSeek-V4 preview in April 2026. He is an author of DeepSeek-V4.1-Flash, a multimodal mixture-of-experts model reported in September 2026.

Now at
DeepSeek

Timeline

  1. January–March 2024

    DeepSeek LLM and DeepSeek-VL

    In January 2024, DeepSeek LLM appears with him among its authors; the project introduces 7B and 67B open-source language models and a dataset of 2 trillion tokens. DeepSeek-VL, an open-source vision-language model, follows in March 2024 with his name on its author list.

  2. May 2024

    The DeepSeek-V2 model

    DeepSeek-V2 counts him among its authors in May 2024. The mixture-of-experts model has 236B total parameters, 21B activated for each token and a 128K-token context length, and it uses Multi-head Latent Attention and DeepSeekMoE.

  3. December 2024

    The DeepSeek-V3 report

    He is among the authors of the DeepSeek-V3 Technical Report in December 2024, which presents a mixture-of-experts language model with 671B total parameters and 37B activated for each token.

  4. September 2025

    DeepSeek-R1 in Nature

    Nature publishes the DeepSeek-R1 paper in September 2025, with him among its authors from the DeepSeek-AI Team in Hangzhou, China. The paper shows that the reasoning abilities of LLMs can be incentivized through pure reinforcement learning, removing the need for human-labelled reasoning trajectories.

  5. October 2025

    Co-authoring DeepSeek-OCR

    With Haoran Wei and Yukun Li, he co-authors DeepSeek-OCR in October 2025, a study of compressing long contexts through optical 2D mapping. The model pairs a DeepEncoder with the DeepSeek3B-MoE-A570M decoder.

  6. December 2025

    DeepSeek-V3.2 and its Speciale variant

    DeepSeek-V3.2 appears in December 2025 with him among its authors. Its high-compute variant, DeepSeek-V3.2-Speciale, reaches gold-medal performance at the 2025 International Mathematical Olympiad and the International Olympiad in Informatics.

  7. April 2026

    The DeepSeek-V4 preview

    The DeepSeek-V4 preview report, submitted in April 2026, lists him among its authors. It presents DeepSeek-V4-Pro with 1.6T parameters (49B activated) and DeepSeek-V4-Flash with 284B parameters (13B activated), both supporting a context length of one million tokens.

  8. September 2026

    DeepSeek-V4.1-Flash and KV caching

    In September 2026, he is among the DeepSeek-AI authors of DeepSeek-V4.1-Flash, a multimodal mixture-of-experts model with 552B backbone parameters and support for contexts of up to one million tokens.

  1. The median h-index is 48 at universities and 14 at frontier AI labsIncludes DeepSeek. The median h-index is 48 at universities and 14 at frontier AI labs. We compare publishing careers, recent citations, dual appointments and job moves across 2,420 university researchers and 308 frontier-lab researchers.
  2. Companies supply the most prior-work hires at 7 of the 9 frontier AI labsIncludes DeepSeek. Our estimates of the organizations where 550 people worked directly before joining 9 frontier AI labs across 620 career transitions, comparing the main feeder organizations per lab.
  3. 5,000 frontier users would be 38.2% of the core labs' 13,100 reported staffIncludes DeepSeek. 5,000 frontier users would be 38.2% of the core labs' 13,100 reported staff. We test the estimate against 3,913 named model contributors and dated all-staff records across Google DeepMind, OpenAI and Anthropic, plus six additional frontier-grade labs.
  4. Google alumni have founded or led 133 other organizations, more than any other AI labIncludes DeepSeek. We estimate that alumni of 17 major AI labs went on to found or lead 350 other organizations, including 226 they founded, and we trace 472 career moves directly between 23 labs since 2010.

Colleagues

People named in this biography, and people with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Yaofeng Sun yet. Send us a free-licensed photo.

  1. [2401.02954] DeepSeek LLM: Scaling Open-Source Language Models with Longtermism arxiv.org. Checked 20 October 2026.
  2. [2403.05525] DeepSeek-VL: Towards Real-World Vision-Language Understanding arxiv.org. Checked 20 October 2026.
  3. [2405.04434] DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model arxiv.org. Checked 20 October 2026.
  4. [2412.19437] DeepSeek-V3 Technical Report arxiv.org. Checked 20 October 2026.
  5. DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learning | Nature nature.com. Checked 20 October 2026.
  6. DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learning - PMC ncbi.nlm.nih.gov. Checked 20 October 2026.
  7. [2510.18234] DeepSeek-OCR: Contexts Optical Compression arxiv.org. Checked 20 October 2026.
  8. [2512.02556] DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models arxiv.org. Checked 20 October 2026.
  9. [2606.19348] DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence arxiv.org. Checked 20 October 2026.
  10. [2609.19969] DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression arxiv.org. Checked 20 October 2026.