PeopleofSuperintelligence

Researcher

Erhang Li

Erhang Li is a DeepSeek-AI researcher whose work spans open language models, reinforcement-learning reasoning, and systems for agent training. His 2024 work includes DeepSeek LLM's 7B and 67B configurations and DeepSeek-V3's 671B-parameter mixture-of-experts model. In 2025, he coauthors DeepSeek-R1 and DeepSeek-V3.2; in 2026, he works on the million-token DeepSeek-V4 and DeepSeek Elastic Compute. As of October 2026, he contributes to DeepSeek-AI model research.

Timeline

  1. 2024

    Scaling DeepSeek LLM

    In 2024, he coauthors DeepSeek LLM, which studies scaling open-source language models in the 7B and 67B configurations. The team uses a 2-trillion-token pretraining dataset, and the 67B Chat model outperforms LLaMA-2 70B on code, mathematics and reasoning benchmarks.

  2. 2024

    DeepSeek-V3 technical report

    In 2024, he coauthors the DeepSeek-V3 report, which describes a 671B-parameter mixture-of-experts model with 37B parameters active per token. It introduces multi-token prediction and auxiliary-loss-free load balancing.

  3. 2025

    DeepSeek-R1 reasoning research

    In 2025, he coauthors DeepSeek-R1 research on reasoning through reinforcement learning. Its published training pipeline combines rejection sampling, reinforcement learning and supervised fine-tuning.

  4. December 2025

    DeepSeek-V3.2 model

    In December 2025, he coauthors DeepSeek-V3.2, which combines efficient sparse attention with scaled reinforcement learning to improve reasoning and agent performance.

  5. April 2026

    DeepSeek-V4 long-context models

    In April 2026, he coauthors the DeepSeek-V4 report, which introduces Pro and Flash mixture-of-experts models that support a one-million-token context.

  6. September 2026

    DeepSeek Elastic Compute

    In September 2026, he coauthors DeepSeek Elastic Compute, a production sandbox platform with function-call, container, microVM and full-VM backends for large-scale agent training.

  7. As of October 2026

    Research contributor at DeepSeek-AI

    As of October 2026, he contributes to DeepSeek-AI model research, including the 2026 DeepSeek-V4 project.

  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 Erhang Li yet. Send us a free-licensed photo.

  1. DeepSeek LLM: Scaling Open-Source Language Models with Longtermism arxiv.org. Checked 19 October 2026.
  2. DeepSeek-V3 Technical Report arxiv.org. Checked 19 October 2026.
  3. DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning arxiv.org. Checked 19 October 2026.
  4. DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learning nature.com. Checked 19 October 2026.
  5. DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models arxiv.org. Checked 19 October 2026.
  6. DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence arxiv.org. Checked 19 October 2026.
  7. DeepSeek Elastic Compute (DSec): A Sandbox Infrastructure for Effective Agentic Training at Scale arxiv.org. Checked 19 October 2026.
  8. Erhang Li | 500AI 500ai.org. Checked 19 October 2026.