I hope it’s K3 level and below 1T, we need more competition!
You must log in or register to comment.
Very likely the same architecture, just with more post training.
What makes you say that? Not arguing, just curious.
Just repeating rumors (sorry, should have been clearer): GLM 5.3 is 5.2 with extra post training, their big upcoming model is 5.5. It kind of makes sense too; pretraining on a new architecture/size is expensive and it’s natural for these companies to try to wring an extra minor version or two out of each one.
glm 5.2 is already pretty good, soo…
quite large though!
on the extreme opposite side of the spectrum:
https://huggingface.co/LiquidAI/LFM2.5-2.6B
or… really extreme… https://huggingface.co/AxiomicLabs/GPT-X2.5-135MAre these any good? I mean, within their size class, obviously - not expecting them to compare to Kimi K3!
Models the size of that 135M one open up some interesting use cases for edge devices.
