AMD and PyTorch ship a playbook to run GPT-OSS-20B inference locally with ROCm
PyTorch · x · 2026-11-22
AMD, cross-posted by the PyTorch account, promotes an AI Playbook for running LLMs on PyTorch with AMD ROCm software.
- Covers the full stack: model, framework, ROCm software, driver, OS, and supported AMD hardware (Ryzen AI APUs/Halo, Radeon GPUs)
- The demo walks through GPT-OSS-20B inference plus a document summarizer, recommending 32GB system memory; validation matrix maps PyTorch 2.13/2.14 to driver versions, last validated October 2026
- Related playbooks include local LLM coding with Qwen3-Coder, n8n automation, fine-tuning with LLaMA Factory and Unsloth, and multi-GPU RCCL clustering
More from coding & agent
- Keep context out of the prompt: RLM-27B beats base Qwen3-27B by 26.9% — dosco · 2026-11-22
- instant_record compiles Rails ActiveRecord models to WASM for instant, offline browser apps — danshipper · 2026-11-22
- Client demoing your AI build internally without prompting is the real adoption signal — aryanXmahajan · 2026-11-22
- Animama: open-source AI desktop companion with VRM avatars and versioned memory files — kavidy · 2026-11-22
- Building an AI team with Claude: an AI employee that already knows its job — blaizedsouza · 2026-11-22
- SteerCamp ships Mac app: delegate work to a crew of AI members on cloud machines — pramodk73 · 2026-11-22