Text Generation
Transformers
PyTorch
English
llama
text-generation-inference
Inference Endpoints

GTPQ, GGUF ver request

#1
by Yhyu13 - opened

@TheBloke
Would you like to make the quantized version of this model for the community. It is current ranked #1 on the LLM leaderboard, and also is the #1 model to surpass 74 avg score

msedge_5zVJJhEqCG.png

Hello,
https://huggingface.co/TheBloke/FashionGPT-70B-V1.1-GGUF

:-)

Edit:
Without holding back enterprising people, or even innovation, I notice that the author doesn't seem very active, or even in a hurry to put information into his models.
==> His first template was created 4 days ago, and is still empty.

This is a problem because we don't know what's inside, or how it was formed (Only FashionGPT ? or a mix with other databases?
For example, if the author includes test data (ARC, HellaSwag, MMLU, TruthfulQA, ...) in his training set, it will logically be stronger, in a biased way of course. Over time, we'll have to see if he stays number 1... or just stays in this ranking.
(I notice that Marcoroni-70b has disappeared. And indeed, I tested it, and it didn't seem any stronger than my Orca_v3_70b).

All right, I'm a pessimist! Let's give him a chance :-)

Yep, its done!

Yhyu13 changed discussion status to closed

Sign up or log in to comment