Kimi K3 is now live

Re:Kimi K3 is now live by orbital-decay on Monday July 20, 2026 @11:36AM (#48980361)

The model is trained to follow the default template pretty closely, breaking it usually results in much worse performance and better output variance. Certain models with synthetic data in pre-training can melt down completely. At some point into this breakage you can just take the base model and it will be better.

If you want to use a custom chat scheme, use it as an overlay, don't break the default chat/tool use/reasoning template.

[ Parent | on HN ]

Re:Kimi K3 is now live by moritzwarhier on Monday July 20, 2026 @03:37PM (#48983834)

Thanks!

Matches my superficial experiments with trying to tweak Ollama's "modelfile" using some LLaMa- or gpt-oss-based instruction-tuned model as "base".

I need to experiment more with base models. The time from the end of 2019 onwards, when I first came across talktotransformer, it felt so magical.

Getting meaningful things out of these things can feel so... restraining.

And on the other hand: I'm tbh freshly stuck in the stage of being amazed at what current frontier coding models and apps can do.

[ Parent | on HN ]