ExLlama is an extremely optimized GPTQ backend for LLaMA models. It features much lower VRAM usage and much higher speeds due to not relying on unoptimized transformers code.

It is highly optimized model loader for GPTQ models. It's an alternative to options like AutoGPTQ or GPTQ-for-LLaMA, and provides faster text generation speeds.

With this update, anyone running GGML exclusively might find some interesting results switching over to a quantized model and testing the changes. I haven't had a chance yet myself, but I will post some of my own benchmarks and results if I find the time for it.

I for one am excited to see the efficiency battles begin. Getting compute down is going to be the most important hurdles to overcome.

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

[–] ArkyonVeil 1 points 1 year ago (1 children)

It would be absolutely awesome, with infinite context length that would mean a much greater ease when it comes to handling models. I can be lazy and instead of creating a LORA, just use an entire book's style as a reference right there in the prompt.

For programmers, just dump the entire codebase, or Documentation.

Of course, all this is only possible if VRAM is less of a bottleneck than it currently is, as well as the fact that it can reliably reference information on an arbitrarily large context. (Not much use having huge context if performance degrades, it loses its marbles or forgets key pieces of information along the way)

[–] Blaed 2 points 1 year ago* (last edited 1 year ago)

I'm with you there. I love how Mosaic just fed the entire Great Gatsby to StoryWriter. This is the sort of context length I need in my life. Would make my projects so much easier. I don't think we're too far from having it on consumer hardware.

You should check out my latest post - which ironically addresses parts of your first comment, but you still need a lot of VRAM... 6000+ tokens context is now possible with ExLlama.

It's crazy to see how fast these developments are happening!