Huh the unsloth Q4_K_XL variant running with 5-6 t/s on a 16GB 9070 XT, 128 GB RAM and an NVME ssd. This is shockingly usable 🤯
Did some more testing on it? This isnt a small model and i havent found the space to save it on my machines yet haha
make sure that holds up with large context, you might need to step down to Q3 (which I've heard is still good for this model, many people are even using IQ2)
Wow, already unsloth quantizations available? That was fast.
Didn't even know they released the weights until just about now.
This will probably also mean that we will have alternative providers on Openrouter soon. Can't wait.
DS V4 will have a huge impact on many usecases. Dirt cheap and insanely capable, if only a part of the benchmarks translate to real life.
Increasingly interesting times for the big US AI companies, I guess...
Anyone knows, why 4bit quant is only marginally smaller than 8bit quant, though?
And shouldn't 8bit be roughly equal to the number of parameters anyway, so ~300GB?
the original model has a lot of parts that were natively trained in 4 bit, so those layers can't go higher
this appears to be the untouched model in GGUF format, 180 GB, MXFP4
https://huggingface.co/bartowski/DeepSeek-V4-Flash-0731-GGUF
Does antirez have new quants for ds4 ready?
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.
Rules:
Rule 1 - No harassment or personal character attacks of community members. I.E no namecalling, no generalizing entire groups of people that make up our community, no baseless personal insults.
Rule 2 - No comparing artificial intelligence/machine learning models to cryptocurrency. I.E no comparing the usefulness of models to that of NFTs, no comparing the resource usage required to train a model is anything close to maintaining a blockchain/ mining for crypto, no implying its just a fad/bubble that will leave people with nothing of value when it burst.
Rule 3 - No comparing artificial intelligence/machine learning to simple text prediction algorithms. I.E statements such as "llms are basically just simple text predictions like what your phone keyboard autocorrect uses, and they're still using the same algorithms since <over 10 years ago>.
Rule 4 - No implying that models are devoid of purpose or potential for enriching peoples lives.