Hugging Face's logo

Join the conversation

Join the community of Machine Learners and AI enthusiasts.

Sign Up
Back to feed
AesSedai 
posted an update 21 days ago
Post
1602
Hi all,

I'm posting this as sort of an informal notice + poll. I'm down to about 700GB free of HF space and there's MiniMax-M3 on the horizon, plus a couple other models I'd like to quant like the Nex-N2 Pro finetune. I've already super-squished all of my quant repositories to free up any LFS space that might have been lingering there, but I'm back near the cap again now.

To free up some space, I'm planning to remove these three older GLM quants:
- GLM-4.5: 1.23TB
- GLM-4.6: 728GB
- GLM-4.7: 787GB

I'm open to other suggestions as well, and I'll wait a few days before removing anything in case someone wants to download a version before I get rid of them.

Thanks!

Edit 06/14/2026:

I've created an organization
QuantPasture and moved the older GLM quants there.

My GLM-4.6-Derestricted has migrated to ReadyArt: ReadyArt/GLM-4.6-Derestricted-GGUF
agentsmit

Hi, I personnaly think glm 4.5

dehnhaide

Those 3 seem to be the right choice. Btw, how much is the next tier storage on HF for you? Maybe we all that enjoy your quants could help you with a bit of crowdfunding?

·
AesSedai

It's a flat $12 per TB per month for public storage. HuggingFace graciously gave me a 10TB storage grant in February, but that's nearly full now. I wouldn't mind asking for another grant, but I wanted to wait at least six months before asking to be polite and a good steward of the space they've already granted me.
Trilogix1

Create a team with new profiles all yours (Aesedai .01, .02...) where you store and organize according to your needs and you keep your models and main profile. Space issue solved, and is a faster solution with many more benefits.

·
AesSedai

I might do that and have a pointer to "old" quants, hmm.
geveent

Thank you for the heads-up. I should probably download them before they are gone.
chimpera

I have have my own copy but GLM-4.7 fits in 256g making it in my mind still relevant. Maybe remove q5? Thanks for you models.
gghfez

For the older Chinese models (not the "de-restricted" GLM-4.6), you could "archive" them to Modelscope.
The free colab instance can download from here -> upload to modelscope one gguf at a time.
Then just put a reference/link in the model cards and squash the repos.
In this post