
Splash
Decodes Qwen3.8-27B and Qwen3.6-35B-A3B faster than llama.cpp on Apple M3 and newer.
Benchmarks
- On M3 Max with Qwen3.8-27B (UD-Q4_K_M), it decodes at 92 tok/s, against 17 tok/s for llama.cpp (5.4×). Source
- On M3 Max with Qwen3.6-35B-A3B (UD-Q4_K_M), it decodes at 209 tok/s, against 66 tok/s for llama.cpp (3.2×). Source
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Something wrong or out of date?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.