mikepapadim

6 ms·

31.
Building libraries for LLM inference in pure modern Java(youtube.com)
2 points by mikepapadim 2y ago|0 comments
32.
Java VectorAPI compatiblity with TornadoVM GPU programming framework(github.com)
2 points by mikepapadim 3y ago|0 comments
33.
Java GPU pre/post processing with ONNX RT and TornadoVM(github.com)
14 points by mikepapadim 3y ago|0 comments
34.
by mikepapadim 3y ago
A comprehensive analysis of the memory behavior of 30 Dacapo and Renaissance Java applications using a dual profiling methodology with NUMAProfiler and PerfUtil in MaxineVM, identifying various memory pressures and JVM impacts.
35.
Multifaceted Memory Analysis of Java Benchmarks with NUMAProfiler and PerfUtil [pdf](pure.manchester.ac.uk)
1 points by mikepapadim 3y ago|1 comments
36.
by mikepapadim 3y ago
How is this related to the $RAD coin?
37.
by mikepapadim 3y ago
https://github.com/mikepapadim/llama-shepherd-cli
38.
Ask HN: Do you know any new llama2.c implementations not mentioned in the repo
1 points by mikepapadim 3y ago|1 comments
39.
by mikepapadim 3y ago
most of the models support the tinyllamas, regarding gguf/ggml and safetensors each implementation has its own model importers, so there is not guarantee that all types can be consumed by all implementations
40.
by mikepapadim 3y ago
Not yet, this is the end-goal of this repo, to be able to do this kind of perf evaluation.
41.
by mikepapadim 3y ago
hello, I am planning to add also a runner option and a benchmarking option among different implementations. This is just an MVP version while trying to keep track of all llama2 implementations as the ones in the original repo is a bit outda
42.
Llama2-shepherd a CLI tool to install multiple implementations of the llama2(github.com)
59 points by mikepapadim 3y ago|8 comments
43.
Java and LLMs: are we there yet?(medium.com)
4 points by mikepapadim 3y ago|0 comments
44.
by mikepapadim 7y ago
Indeed aparapi worked by converting java to OpenCL. However, exposed to the user many aspects of GPU programming such as thread indexing (e.g global ids) and memory allocation for specific optimizations (e.g local memory). In the case of To