New York City

Oleksandr Kuvshynov

Expert
@okuvshynov
okuvshynov.github.io/

slowllama. Finetune llama2-70b and codellama on MacBook Air without quantization

449

b63. Micro-benchmarking library for C and C++ with PMU counters tracking

63

llama-sandbox. A collection of experiments related to LLM inference with llama.cpp/mlx

40

cubestat. Horizon chart for CPU/GPU/Neural Engine utilization monitoring. Supports Apple M1-M4, Nvidia GPUs, AMD GPUs

28

flamereport. FlameGraph visualization for terminal

11

fewlines. Create histograms, time series charts and dashboards in command-line output and log files.

10

qna. AI pretends to be paper/textbook author, you can ask it questions about the paper as a whole, specific parts of it right in the PDF viewing app (e.g. Apple Preview) using annotations and see the replies there.

6

vimqq. local LLM chat for vim with dynamic warmup, forks and multi-model support

3

dotfiles. Lua

2

experiments. C++

2

golem. Python

2

bioproxy. Go

1

rlscout. In progress attempt to use Deep RL to solve some game

1
13
Apply