Skip to main content

Add DeepSeek R1 671B

It would be cool to have the 671B parameter model with its web search capablities, rather than the 70B thats only for speed.

Status: Completed4 comments

Log in to comment and vote

Comments4

  • Amber Flashlight

    •

    Feb 1, 2025

    this was added today, but not to the API yet…

  • Azure Leaf

    •

    Jan 30, 2025

    i cannot stand that the team still put censored or restricted modele,i think they should allowed users to prompts or personalise their own deepseek for now you cannot but i’m sure the team will work on it pretty soon.

  • Apricot Koala

    •

    Jan 30, 2025

    Ideally any Deepseek model should be first uncensored and limiters removed:

    https://huggingface.co/huihui-ai/DeepSeek-R1-Distill-Qwen-14B-abliterated-v2

  • Yellow Nucleon

    •

    Jan 28, 2025

    we need Deepseek R1 671B, as well as 70B with 128k context length