#rocm on Hacker News

6 stories tagged rocm, newest first.

Screenshot of dwarfstar.sh: From the creator of Redis; run LLM locally with ds4

From the creator of Redis; run LLM locally with ds4 (dwarfstar.sh)

332 points by fibo 23 days ago | 95 comments

Landing page for DwarfStar 4 (ds4), an MIT-licensed C inference engine for running open-weight LLMs like DeepSeek V4, GLM 5.x and Qwen locally on Mac, CUDA and ROCm.

local-llm-inference ds4 deepseek c

Screenshot of github.com: CuPy: NumPy and SciPy for GPU

CuPy: NumPy and SciPy for GPU (github.com)

347 points by tanelpoder 2 years ago | 97 comments

GitHub repository for CuPy, a NumPy- and SciPy-compatible array library for GPU-accelerated computing in Python with CUDA and ROCm support.

cupy python gpu numpy

Screenshot of ollama.com: Ollama now supports AMD graphics cards

Ollama now supports AMD graphics cards (ollama.com)

578 points by tosh 2 years ago | 194 comments

Ollama blog post from March 14, 2024 announcing preview support for running models on AMD graphics cards.

ollama amd gpu rocm

Screenshot of cgchannel.com: Open-source project ZLUDA lets CUDA apps run on AMD GPUs

Open-source project ZLUDA lets CUDA apps run on AMD GPUs (cgchannel.com)

357 points by drakerossman 2 years ago | 133 comments

CG Channel news article about ZLUDA, an open-source project that lets unmodified CUDA applications run on AMD GPUs.

zluda cuda amd gpu

Screenshot of phoronix.com: AMD funded a drop-in CUDA implementation built on ROCm: It's now open-source

AMD funded a drop-in CUDA implementation built on ROCm: It's now open-source (phoronix.com)

945 points by mfiguiere 2 years ago | 360 comments

Phoronix article reporting that AMD quietly funded ZLUDA, a drop-in CUDA implementation running on ROCm, which has now been released as open source.

zluda cuda rocm amd

Screenshot of hpcwire.com: AMD may get across the CUDA moat

AMD may get across the CUDA moat (hpcwire.com)

498 points by danzheng 3 years ago | 266 comments

HPCwire article by Doug Eadline discussing how AMD's Instinct GPUs and software stack might overcome Nvidia's CUDA software lock-in for GenAI workloads.

amd cuda nvidia gpu

#rocm on Hacker News

From the creator of Redis; run LLM locally with ds4(dwarfstar.sh)

CuPy: NumPy and SciPy for GPU(github.com)

Ollama now supports AMD graphics cards(ollama.com)

Open-source project ZLUDA lets CUDA apps run on AMD GPUs(cgchannel.com)

AMD funded a drop-in CUDA implementation built on ROCm: It's now open-source(phoronix.com)

AMD may get across the CUDA moat(hpcwire.com)