ZOYKU · Open Source Intelligence · Repository archive
youssofal/mtplx
The fastest way to run Qwen 3.8 Flash Next, Qwen 3.8 27B and Ternary Bonsai 2 27B on a Mac: 125 tok/s in OpenCode on an M5 Max, and a 27B model on 16 GB Macs. Native MTP speculative decoding on Apple Silicon, exact at any temperature. OpenAI and Anthropic compatible local server.
PythonApache-2.0Actively maintained · Pushed 1d ago#anthropic-compatible#apple-silicon#claude-code#flash-next#inference-engine#llm-inference#local-ai#local-llm
stars
2.7k
forks
211
issues
69
Daily star gain
—
days old
2d
6-Axis Combat Radar
Combat Limits & Architectural Persona
Architectural Persona
Specialized Tactical Vanguard
Distinct edge and dedicated footprint in specialized territory
Combat Score
56
/ 100
7-Day Star Velocity
0
Fork/Star Adoption
7.9%
Cumulative stars
Next Milestone: 5k (2.3k stars to go)+1 (2d)2026-11-08: 2.7k
Daily star gain
Max: +1 / day
Similar projects · shared language & topics
Maintainer Badge Studio
Embed GitHub README Live Badge
Embed dynamic SVG badges into your README to reflect real-time rankings and documentation mirrors.
Preview240 x 64 px SVG
GitHub Dark Theme
GitHub Light Theme