News for hackers. Stuff that matters.

A comment on DeepSeek-R1 with Dynamic 1.58-bit Quantization. Parent comment

Re:DeepSeek-R1 with Dynamic 1.58-bit Quantization by whimsicalism on Tuesday January 28, 2025 @10:16AM (#42853253)
no they are not, they intentionally remove every reference to this not being r1 from the cli and changed the names from the ones both Deepseek and Huggingface used.
[ Parent | on HN ]
Re:DeepSeek-R1 with Dynamic 1.58-bit Quantization by nyrikki on Tuesday January 28, 2025 @04:06PM (#42858004)
They used short strings for the names, which is very different from deception.

https://ollama.com/search

> DeepSeek's first-generation of reasoning models with comparable performance to OpenAI-o1, including six dense models distilled from DeepSeek-R1 based on Llama and Qwen.

Well I guess if you are in the Enterprise Java naming model you would expect something like "­Visitor­Model­Utils­List­Getter­Adapter­Interceptor­Message­Manager­Driven­Observer­Pool"

If you look at their API docs you will see:

    model: name of the model to push in the form of <namespace>/<model>:<tag>
I don't think there is any reason to jump to the conclusion it is some type of conspiracy here, just naming things based on a API that probably didn't think about distillation when they created it.
[ Parent | on HN ]
Re:DeepSeek-R1 with Dynamic 1.58-bit Quantization by blacklightpy on Thursday January 30, 2025 @11:55AM (#42879617)
Yet, I did not see a single issue made on the GitHub repository, and I just made one myself (https://github.com/ollama/ollama/issues/8698).
[ Parent | on HN ]