News for hackers. Stuff that matters.

A comment on Qwen2.5-VL-32B: Smarter and Lighter. Parent comment

Re:Qwen2.5-VL-32B: Smarter and Lighter by theturtletalks on Monday March 24, 2025 @03:56PM (#43464802)
Wouldn’t they just run R1 locally and not have any censorship at all? The model isn’t censored at its core, it’s censored through the system prompt. Perplexity and Huggingface have their own versions of R1 that is not censored.
[ Parent | on HN ]
Re:Qwen2.5-VL-32B: Smarter and Lighter by hmottestad on Monday March 24, 2025 @04:10PM (#43464929)
I tried R1 through Kagi and it’s similarly censored. Even the distill of llama running on Groq is censored.
[ Parent | on HN ]
Re:Qwen2.5-VL-32B: Smarter and Lighter by theturtletalks on Monday March 24, 2025 @05:22PM (#43465527)
Kagi may be using the official DeepSeek API and not hosting the model itself. There is work being done to make it completely uncensored:

https://github.com/huggingface/open-r1

https://ollama.com/huihui_ai/deepseek-r1-abliterated

I was mistaken though, it is more than just a system prompt causing the censorship.

[ Parent | on HN ]
Re:Qwen2.5-VL-32B: Smarter and Lighter by hmottestad on Tuesday March 25, 2025 @05:38AM (#43469442)
Kagi uses R1 through Fireworks.ai, Together.ai and Groq.

https://help.kagi.com/kagi/ai/llms-privacy.html

[ Parent | on HN ]