Kalsa
Engines / SharpAI / SwiftLM

SwiftLM

Apple M1–M6EngineMIT

Streams experts from the SSD to cut GPU memory use on Apple Silicon Macs.

What it does better

Streaming the experts from the SSD cuts peak GPU memory for Qwen3.6-35B-A3B 4-bit from 20.2 GB to 5.7 GB at a prompt of about 550 tokens, on a base 32 GB Mac mini with an M6

Trade-off: decode falls from 48.4 to 14.1 tok/s and prefill from 687 to 263 tok/s.

Source

Benchmarks

The author hasn't published benchmarks yet.

Listing

Listed: Passed all seven checks.

Security review

Not reviewed yet.

Install

To install it, follow the project's own instructions on GitHub. The engine runs with your user's permissions. macOS doesn't sandbox it, so install only engines you trust.

Something wrong or out of date? Write to help@kalsa.io, or ask us to remove this page.