News for hackers. Stuff that matters.
Re:A Competitor Crippled a $23.5M Bootcamp by Becoming a Red... by trissi1996 on Thursday October 09, 2025 @04:20AM (#45524967)
You're just wrong.

Vibes are not a meaningful comparison.

You can't compare different quantizations(Q4_K_M & Q4_K_XL) and also use a temp of 0.7 and use that for any meaningful comparison.

All difference you notice could be either due to temp randomness or quantization differences even if the original weights are exactly the same.

Also these were likely sourced differently so the quantizers could've used different quantization importance matrizes or process.

The weights are the same, the weights are all that matters.

Re:A Competitor Crippled a $23.5M Bootcamp by Becoming a Red... by trilogic on Thursday October 09, 2025 @04:40AM (#45525082)
Nonsense, there is a big difference of tokens generated in the output. I know my stuff, and I can recognize Deepseek reasoning and Qwen thinking. Also I explained how agentic is working in the models making it very different. The quantization is irrelevant to tokens generated and agentic behavior in a model. Say what you want, you or anyone else can´t change the facts.