Any plans for more Dolphin/Mistral models with LASER optimization?

#13
by breimers - opened

I have found the performance of dolphin-2.6-mistral-7b-dpo-laser quantized models to be among the very best of the open-source models out there. With less than 8GB of memory I can run a full fledged LLM.

Would love to know if you have future plans to continue LASER optimization of your models because this is awesome.

deleted
This comment has been hidden
deleted
This comment has been hidden

Hi... sorry for the delay response.
Join our discord server. It would be a pleasure talking to you.

And yes, we will release newer versions of the laser model.
Have you tried laserxtral?

fernandofernandes changed discussion status to closed

Thank you! Gonna join the discord

Sign up or log in to comment