New Alchemy Studio

DeepSeek-R (R1)

DeepSeekChinaOpen sourceOpen weightsmittext

Reasoning models trained with RL.

Release history (12)

  1. DeepSeek-R1-0528-Qwen3-8B
    stableMay 29, 2025
    deepseek-ai/DeepSeek-R1-0528-Qwen3-8B

    Detected on Hugging Face Hub.

  2. DeepSeek-R1-0528
    stableMay 28, 2025
    deepseek-ai/DeepSeek-R1-0528

    Detected on Hugging Face Hub.

  3. DeepSeek-R1
    stableJan 20, 2025
    deepseek-ai/DeepSeek-R1671B params

    Open-weights reasoning model, MIT.

  4. DeepSeek-R1 (distill)
    stableJan 20, 2025
    deepseek-ai/DeepSeek-R1-Distill-Llama-70B70B params

    Reasoning distillations.

  5. DeepSeek-R1-Distill-Qwen-7B
    stableJan 20, 2025
    deepseek-ai/DeepSeek-R1-Distill-Qwen-7B

    Detected on Hugging Face Hub.

  6. DeepSeek-R1-Distill-Qwen-1.5B
    stableJan 20, 2025
    deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B

    Detected on Hugging Face Hub.

  7. DeepSeek-R1-Distill-Qwen-32B
    stableJan 20, 2025
    deepseek-ai/DeepSeek-R1-Distill-Qwen-32B

    Detected on Hugging Face Hub.

  8. DeepSeek-R1-Distill-Qwen-14B
    stableJan 20, 2025
    deepseek-ai/DeepSeek-R1-Distill-Qwen-14B

    Detected on Hugging Face Hub.

  9. DeepSeek-R1-Zero
    stableJan 20, 2025
    deepseek-ai/DeepSeek-R1-Zero

    Detected on Hugging Face Hub.

  10. DeepSeek-R1-Distill-Llama-8B
    stableJan 20, 2025
    deepseek-ai/DeepSeek-R1-Distill-Llama-8B

    Detected on Hugging Face Hub.

  11. DeepSeek-R1 (preview)
    deprecatedNov 20, 2024
    deepseek-ai/DeepSeek-R1-Lite-Preview671B params

    Reasoning preview.

  12. DeepSeek-R2
    unconfirmed
    deepseek-ai/DeepSeek-R2671B params

    Next-gen reasoning. [Flagged: pre-release at time of seed; confirm before publishing.]

Spotted something out of date? Edit in admin.