Pipe Network PIPE: Kimi K3 MLX 포트 릴리스
Pipe Network은 7월 28일에 Kimi K3용 오픈 소스 MLX 포트를 출시했습니다. 스트리밍 변환기와 REAP 가지치기를 통해 모델 크기가 1.6TB에서 약 350GB로 줄어들어 Mac Studio에서도 실행할 수 있게 되었습니다.
이밴트: 2026년 7월 28일 UTC
Pipe Network
@pipenetwork
@pipenetwork
Run Kimi K3 on a Mac Studio 🫰
K3 is 2.8T parameters and 1.6TB on disk, which makes it impossible to run on Apple Silicon.
Until now.
Our MLX port is now open source: https://github.com/PipeNetwork/kimi-k3-mlx
To accomplish this, we solved two things:
1. We wrote a streaming converter that walks one layer at a time, so that mlx_lm doesn't need to materialize the whole model.
2. REAP pruning sits on top and scores all 896 experts against a calibration corpus to keep only ones your workload needs.
That's what brings K3 down to 350GB and inside a Mac Studio.
K3 is 2.8T parameters and 1.6TB on disk, which makes it impossible to run on Apple Silicon.
Until now.
Our MLX port is now open source: https://github.com/PipeNetwork/kimi-k3-mlx
To accomplish this, we solved two things:
1. We wrote a streaming converter that walks one layer at a time, so that mlx_lm doesn't need to materialize the whole model.
2. REAP pruning sits on top and scores all 896 experts against a calibration corpus to keep only ones your workload needs.
That's what brings K3 down to 350GB and inside a Mac Studio.
Pipe Network
@pipenetwork
@pipenetwork
Get the full recipe, converter, calibration and pruning scripts here: https://t.co/q3jf1q3GZG
Quants published on Hugging Face:
1. REAP73: https://t.co/FvuDPBAL6B
2. REAP80: https://t.co/DQx7he4jex
3. REAP73 specialized in Chinese + code: https://t.co/woJkmKhu2B
Quants published on Hugging Face:
1. REAP73: https://t.co/FvuDPBAL6B
2. REAP80: https://t.co/DQx7he4jex
3. REAP73 specialized in Chinese + code: https://t.co/woJkmKhu2B
이벤트가 곧 시작됩니다
0
D
0
H
0
M
0
S
Y
28 7 16:38 (UTC)
✕
✕