Articles
- Occasionally, I write about LLM AI technologies that interest me on this blog.
- If you want to begin your learning journey about LLM AI, this blog will be very helpful for you.
Projects
-
LLM semantic search — 22.08.2025
-
PyTorch pp512 and tg128 LLM Benchmark — 27.08.2025
General
-
Begin your journey in LLM — 14.08.2025
-
NVIDIA CUDA Compatibility — 14.08.2025
-
CUDA & PyTorch Compatibility by Compute Capability — 14.08.2025
-
TensorFlow-friendly causal LMs in Transformers 4.x — 18.08.2025
-
NVIDIA CUDA Docker Base Images — 23.08.2025
-
NVIDIA CUDA Docker Images for PyTorch / TensorFlow — 23.08.2025
-
AMD ROCm Docker Base Images — 23.08.2025
-
AMD ROCm Docker Images for PyTorch / TensorFlow — 23.08.2025
-
BitsAndBytes CUDA Compatibility — 28.08.2025
-
BitsAndBytes ROCm Compatibility — 28.08.2025
-
NVIDIA Container Toolkit for Docker — 03.09.2025
-
FlashAttention compatibility — 25.12.2025
-
PyTorch dtype compatibility — 05.01.2026
GPU
-
NVIDIA Tesla M10 GPU — 05.07.2025
-
NVIDIA Tesla K80 GPU — 02.08.2025
-
AMD Instinct Mi50 GPU — 07.08.2025
-
NVIDIA Tesla V100 GPU SXM2 — 03.09.2025
-
Apple M1 16Gb Unified RAM — 09.09.2025
-
NVIDIA Tesla P100 GPU — 28.09.2025
-
NVIDIA RTX 3090 GPU — 23.12.2025
Software
-
Mistral 7b CUDA low memory GPU PyTorch Test — 14.06.2025
-
Stable Diffusion v1.5 CUDA low memory GPU PyTorch Test — 25.06.2025
-
Mistral 7b ROCm PyTorch Test — 08.08.2025
-
Stable Diffusion v1.5 ROCm PyTorch Test — 09.08.2025
-
llama.cpp - run LLM everywhere — 10.08.2025
-
TensorFlow vs PyTorch GPT2 — 18.08.2025
-
AMD ROCm PyTorch in Docker Test — 24.08.2025
-
AMD ROCm TensorFlow in Docker Test — 25.08.2025
-
Vulkan llama.cpp in Docker Test — 26.08.2025
-
Compilation PyTorch BitsAndBytes for CUDA 11.4 — 28.08.2025
-
ROCm PyTorch BitsAndBytes Mistral Test — 29.08.2025
-
Compilation BitsAndBytes for ROCm 6.2 — 29.08.2025
-
Stable Diffusion v1.5 ROCm BitsAndBytes PyTorch Test — 30.08.2025
-
SUNO Bark ROCm PyTorch Test — 31.08.2025
-
NVIDIA CUDA PyTorch in Docker Test — 03.09.2025
-
NVIDIA CUDA TensorFlow in Docker Test — 04.09.2025
-
Mistral 7b General VS Instruct Test — 29.09.2025
-
Stable Diffusion 1.5 LoRA Tranning Test — 30.09.2025
-
CUDA PyTorch BitsAndBytes FlashAttention2 Mixtral-8x7B Test — 26.12.2025
-
Compilation FlashAttention for CUDA 12.8 — 26.12.2025
-
Stable Diffusion 1.5 vs 2.0 vs XL Test — 27.12.2025
-
Deterministic Test with CUDA PyTorch Mixtral-Small-22b — 28.12.2025
-
WAN 2.1 1.3b diffusers CUDA PyTorch Test — 30.12.2025
-
WAN 2.1 1.3b diffusers CUDA PyTorch BNB4 Test — 02.01.2026
-
Falcon 7b CUDA PyTorch Test — 09.01.2026
-
PyTorch Transformers repetition loop Fix — 09.01.2026
-
GLM4 9B CUDA PyTorch Test — 11.01.2026
-
Olmo3 7B CUDA PyTorch Test — 11.01.2026
-
Yi 9b CUDA PyTorch Test — 11.01.2026
-
Mistral 7b CUDA PyTorch Test — 12.01.2026
-
Open (Fake) Llama 7b CUDA PyTorch Test — 12.01.2026
-
Qwen 8B CUDA PyTorch Test — 13.01.2026
Known issues / errors
-
Due to a serious vulnerability upgrade torch to v2.6 — 30.08.2025
-
PyTorch convert binary weights from bin to safetensors — 01.09.2025
Failed test
- SUNO Bark voice clone ROCm PyTorch Test — 31.08.2025