Michael-A-Kuykendall/shimmy
View on GitHub ↗ ★ 6,015 Rust
⚡ Pure-Rust WebGPU inference engine — OpenAI-API compatible, GGUF native, runs on any GPU. No Python. No llama.cpp. Single binary.
6 mentions on Hacker News by 1 person: 3 stories, 3 comments · first 2025-09-04 · latest 2025-10-09
20082026
-
…collection make it accessible to developers who just want to run large models locally. Release: https://github.com/Michael-A-Kuykendall/shimmy/releases/tag/... Models: https://huggingface.co/MikeKuykendall Happy to answer questions about…
comment by MKuykendall · 2025-10-09 · in Shimmy v1.7.0: Running 42B Moe Models on Consumer GPUs with 99.9% VRAM Reduction -
▲ 3
Shimmy v1.7.0: Running 42B Moe Models on Consumer GPUs with 99.9% VRAM Reductionstory by MKuykendall · 2025-10-09 · 1 comment
-
Side note I built a 2500 star and climbing rust application from stem to stern using it https://github.com/Michael-A-Kuykendall/shimmy
comment by MKuykendall · 2025-09-20 · in Show HN: Rustchain – Rust toolchain AI agent framework universal transpilation -
▲ 2
Show HN: 5MB Rust binary that runs HuggingFace models (no Python)story by MKuykendall · 2025-09-10 · 2 comments
-
…just to run a 4GB model locally. Quick demo - working VSCode + local AI in 30 seconds: curl -L https://github.com/Michael-A-Kuykendall/shimmy/releases/late... ./shimmy serve # Point VSCode/Cursor to localhost:11435 The technical…
comment by MKuykendall · 2025-09-04 · in Show HN: Shimmy – 5MB privacy-first, local alternative to Ollama (680MB) -
▲ 14
Show HN: Shimmy – 5MB privacy-first, local alternative to Ollama (680MB)story by MKuykendall · 2025-09-04 · 13 comments