LocalLLaMA

2881 readers

45 users here now

Welcome to LocalLLaMA! Here we discuss running and developing machine learning models at home. Lets explore cutting edge open source neural network technology together.

Get support from the community! Ask questions, share prompts, discuss benchmarks, get hyped at the latest and greatest model releases! Enjoy talking about our awesome hobby.

As ambassadors of the self-hosting machine learning community, we strive to support each other and share our enthusiasm in a positive constructive way.

founded 2 years ago

MODERATORS

Exllama V2 released! Available in Ooba! Big speed upgrades! (github.com)

submitted 2 years ago* (last edited 2 years ago) by noneabove1182 to c/localllama

3 comments fedilink hide all child comments

Linked is the new repo, it's still in relatively early stages but does work.

I'm using it in oobabooga text-gen-ui and the OLD GPTQ format, so not even the new stuff, and on my 3060 I see a genuine >200% increase in speed:

Exllama v1

Output generated in 21.84 seconds (9.16 tokens/s, 200 tokens, context 135, seed 1891621432)

Exllama v2

Output generated in 6.23 seconds (32.10 tokens/s, 200 tokens, context 135, seed 313599079)

Absolutely crazy, all settings are the same. And it's not just a burst at the front, it lasts:

Output generated in 22.40 seconds (31.92 tokens/s, 715 tokens, context 135, seed 717231733)

And this is using the old format, exllama v2 includes a new way to quant, allowing for much more granular bitrates.

Turbo went with a really cool approach here, you set a target bits per weight, say, 3.5, and it'll automatically adjust the appropriate weights to the appropriate quant levels to achieve maximum performance where it counts, saving data in important weights and sacrificing more on non important ones, very cool stuff!

Get your latest oobabooga webui and start playing!

https://github.com/oobabooga/text-generation-webui

https://github.com/noneabove1182/text-generation-webui-docker

Some models in the new format from turbo: https://huggingface.co/turboderp

you are viewing a single comment's thread
view the rest of the comments

[–] [email protected] 3 points 2 years ago (1 children)

Why'd you create your own dockerfile repo vs just improving/changing the one in the main ooba repo?

[–] noneabove1182 2 points 2 years ago

Good question, at the time I made it there wasn't a good option, and the one in the main repo is very comprehensive and overwhelming, I wanted to make one that was straight forward and easier to digest to see what's actually happening