Magnitude: Releases and Overview
Open-source inference engine written in Rust (Apache-2.0) that compiles and tunes its kernels on the device it runs on. The project reports up to 2x faster than llama.cpp on Metal and CUDA; in our CPU-only measurement it was about 21x slower.
Category: Inference engines and runtimes. Part of our engine and tool tracker.
At a Glance
| Item | Value |
|---|---|
| Repository | magnitudedev/magnitude |
| Latest release | @magnitudedev/cli@0.2.17 (2026-11-09) |
| Install / docs | official documentation |
Release History
Compiled by Local Model Watch from the project’s GitHub releases. Pre-releases are omitted. Where we wrote an article about a release, it is linked in the last column; smaller releases are tracked here only.
Articles on Local Model Watch
- 2026-11-17 Magnitude CLI v0.2.17 Released with Dynamic Context Scaling
- 2026-11-17 Magnitude v0.2.16 Released: Faster Parallel Requests and Docker
- 2026-11-17 Magnitude CLI 0.2.14 Released: Faster CPU Decoding and More
- 2026-11-17 Magnitude @magnitudedev/cli@0.2.12 Released
- 2026-11-17 Magnitude CLI 0.2.11 Released with Custom GGUF Support
- 2026-11-17 Magnitude CLI 0.2.10 Released with Memory Fixes and New Quantizations
- 2026-11-17 Magnitude @magnitudedev/cli@0.2.9 Released
- 2026-11-17 Magnitude CLI 0.2.8 Released with VRAM Fixes and Gemma 4 Support
- 2026-11-17 Magnitude CLI 0.2.7 Released: Major Speedups and Fixes
- 2026-10-06 magnitudedev/magnitude v0.2.6 Released: Major Mac Speedups
- 2026-10-03 Magnitude CLI v0.2.5 Released: M5 Mac & MoE Speedups
- 2026-10-03 Magnitude 0.2.4 Released: Faster Inference and Lower Memory
Last updated 2026-11-17 (JST). Facts above come from the GitHub API; the one-paragraph summary was written by the site’s operator with the help of AI.