-
This is actually spanning both Qibojit and Qibo itself, but being specific for backends, I decided to avoid polluting Qibo's tracker.
It is only a proposal and definitely *not urgent*. The goal is …
-
Hi!
Our team is currently developing sparse symmetric and non-symmetric solvers in JAX (implicitly restarted arnoldi and lanzcos). I have it from hearsay that there are efforts to support this nati…
-
I am bringing `dbcsr` to Macports, where we support all range of macOS including old ones (at least 10.5+).
Two tests fail on 10.6.8 Rosetta (I cannot test native PPC at the moment, away from PPC har…
-
With each token processed the inference speed slows down a little bit, starts to become noticeable at around 50 tokens on 40B Q3_K and adds up.
-
So this is a pretty immense task and I'd start with #45, but...
> RWKV is an RNN with Transformer-level LLM performance, which can also be directly trained like a GPT transformer (parallelizable). …
-
Right now, there's `AdditiveSemigroup` and `MultiplicativeSemigroup` which are then combined into `Semiring`. This is all well and good, but unfortunately there's a huge amount of code out there writt…
-
Just did a very simple run with llama-7b-4bit. It... took a while. Had it run in a screen. But, it worked!
```
root@FriendlyWrt /s/o/llama.cpp (master)# time ./main --color -m models/ggml-model-q4…
-
Reading the current spec it seems that NNEF supports only fixed (constant) shape sizes. Is that correctly understood or not and if so do you intend to support dynamic/variable shapes?
botev updated
5 years ago
-
much of our implementation work to this point has focused on explicit time advance (ET). however, we also want to support implicit timestepping (IT).
ET is driven by a large matrix vector product. …
-
Hello!
In order to properly benchmark the preprocessing and online phase of a certain comparison protocol for linear secret sharing scheme-based MPC protocols (see old draft [here](https://eprint.i…
waamm updated
1 month ago