PeopleofSuperintelligence

Researcher

Guangbo Hao

Guangbo Hao is a coauthor of DeepSeek's language model papers, from DeepSeek LLM in January 2024 to DeepSeek-V4.1-Flash in September 2026. In 2024, he coauthored the Mixture-of-Experts models DeepSeek-V2 and DeepSeek-V3; DeepSeek-V3 has 671B parameters and needed 2.788M H800 GPU hours for its full training. With the DeepSeek-AI Team in Hangzhou, he coauthored DeepSeek-R1, a 2025 study of reasoning learned through pure reinforcement learning, which Nature published in September 2025. His later papers covered DeepSeek-V3.2 and its DeepSeek Sparse Attention in December 2025 and the million-token DeepSeek-V4 series in April 2026.

Timeline

  1. January 2024

    The DeepSeek LLM project

    In January 2024, he coauthors DeepSeek LLM, a DeepSeek-AI project that studies scaling laws for open-source language models in 7B and 67B configurations. The team pre-trains on a dataset of 2 trillion tokens.

  2. May 2024

    DeepSeek-V2 mixture of experts

    He coauthors DeepSeek-V2 in May 2024, a Mixture-of-Experts language model with 236B total parameters, of which 21B are active for each token. Its Multi-head Latent Attention compresses the key-value cache into a latent vector.

  3. December 2024

    The DeepSeek-V3 technical report

    In December 2024, he coauthors the DeepSeek-V3 Technical Report on a 671B-parameter Mixture-of-Experts model that activates 37B parameters per token. DeepSeek-V3 trains on 14.8 trillion tokens and needs 2.788M H800 GPU hours for its full training.

  4. January 2025

    The DeepSeek-R1 reasoning paper

    He coauthors DeepSeek-R1, posted in January 2025. The paper shows that pure reinforcement learning can incentivize reasoning in large language models, removing the need for human-labeled reasoning trajectories.

  5. September 2025

    DeepSeek-R1 in Nature

    In September 2025, Nature publishes the DeepSeek-R1 paper with him among its authors from the DeepSeek-AI Team in Hangzhou, China. Its reinforcement-learning framework yields reasoning patterns such as self-reflection, verification and dynamic strategy adaptation.

  6. December 2025

    DeepSeek-V3.2 and sparse attention

    In December 2025, he coauthors DeepSeek-V3.2, which introduces DeepSeek Sparse Attention for long-context efficiency. Its high-compute variant, DeepSeek-V3.2-Speciale, reaches gold-medal performance at the 2025 International Mathematical Olympiad and International Olympiad in Informatics.

  7. April 2026

    Million-token DeepSeek-V4 models

    He coauthors the preview of the DeepSeek-V4 series in April 2026. Its two Mixture-of-Experts models, DeepSeek-V4-Pro with 1.6T parameters and DeepSeek-V4-Flash with 284B parameters, both support one-million-token contexts.

  8. September 2026

    KV cache compression in V4.1-Flash

    In September 2026, he coauthors DeepSeek-V4.1-Flash, a multimodal Mixture-of-Experts model with 552B backbone parameters and contexts of up to one million tokens. Its design cuts the global KV cache footprint to 890 bytes per token.

  1. Google had the most people shaping AI every year from 2014 to 2025Our estimates of which organizations employed the most people shaping AI from 1961 to 2025, with a year-by-year chart.
  2. The United States employs an estimated 53% of people in our data with a known employment countryOur estimates of where AI careers are concentrated now and how countries’ shares of dated positions changed over time.
  3. 46% of living people in our data with known birth and employment countries work outside their birth countryOur estimate of cross-border careers in AI, from country of birth and the country of each current organization.
  4. More people shaping AI studied at Massachusetts Institute of Technology than at any other institutionOur estimates of where the people shaping AI studied and where their careers led, from education records in cited biographies.

Sources and credits

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

  1. [2401.02954] DeepSeek LLM: Scaling Open-Source Language Models with Longtermism arxiv.org. Checked 19 October 2026.
  2. [2405.04434] DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model arxiv.org. Checked 19 October 2026.
  3. [2412.19437] DeepSeek-V3 Technical Report arxiv.org. Checked 19 October 2026.
  4. Paper page - DeepSeek-V3 Technical Report huggingface.co. Checked 19 October 2026.
  5. [2501.12948] DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning arxiv.org. Checked 19 October 2026.
  6. DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learning | Nature nature.com. Checked 19 October 2026.
  7. DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learning - PMC pmc.ncbi.nlm.nih.gov. Checked 19 October 2026.
  8. [2512.02556] DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models arxiv.org. Checked 19 October 2026.
  9. [2606.19348] DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence arxiv.org. Checked 19 October 2026.
  10. [2609.19969] DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression arxiv.org. Checked 19 October 2026.