-
We need to support AVX-512 instructions in order to support Knights Landing (KNL) and future Xeon processors effectively.
AVX-512 is not monolithic (see [Wikipedia](https://en.wikipedia.org/wiki/AVX-…
-
### What is the problem this feature will solve?
As I am looking into ways to improve astropy performance, I can see that using the new'ish compiler feature of function multi-versioning should improv…
-
This task is to identify potential opportunities to use `Vector512` in these libraries(ASCII/UTF) and add `Vector512 `paths where possible to further accelerate using SIMD.
@dotnet/avx512-contrib …
-
I am interested in selecting a key-value DB and ran the comparison to understand the performance differences.
Didn't expect such big difference between [pebble](https://github.com/cockroachdb/pebble…
-
### Steps to reproduce the issue
```console
$ spack install openmpi@4.1.4 %gcc@7.3.0 +legacylaunchers +gpfs +pmi schedulers=slurm >> log.openmpi 2>&1
```
### Error message
Error message
==> In…
-
### Your current environment
```text
The output of `python collect_env.py`
OS: Ubuntu 24.04.1 LTS (x86_64)
GCC version: (Ubuntu 13.2.0-23ubuntu4) 13.2.0
Clang version: Could not collect
CMake …
-
I have a desktop SKX, model W-2123, which the cpuid code identifies as haswell (obvious with `configure auto`).
It turns out that it doesn't report avx512 vpus due to not parsing the model name. I…
-
### Describe the issue
Snapdragon and Dimensity and every newer devices are using the much much faster ARMv9.2
Or 9.3 or 9.4.
https://en.wikipedia.org/wiki/ARM_architecture_family
https://www.you…
-
### What is the issue?
It's again the https://github.com/ollama/ollama/issues/6011 issue.
**The issue is with embedding call with the model converted using convert_hf_to_gguf.py.**
litellm.ll…
-
I've created a version of the direct sgemm code for AVX2 (it's shared with the AVX512 code with very limited ifdefs, so can compile from the same source).
Question is if and how to integrate.
The …