硬體 / UNIFIED MEMORY

MacBook Pro M5 Max · 64 GB

A portable Apple-native tier for larger quantized models. Sustained inference still depends on thermals, runtime support, and context length.

一手來源

Current Apple notebook64 GB

01 / 實際上限

Qwen3.8 Flash Next REAP-288

清單中排最前的模型可能是勉強適配或需要卸載的。日常使用建議選第一個標為 餘量充足 的結果，並保持上下文適中。

02 / 預設推論服務

MLX LM

Native Apple silicon inference, experimentation and fine-tuning. Only for Apple silicon. Use an MLX-converted model and tune KV cache size for long contexts. The bundled mlx_lm.server is an OpenAI-compatible HTTP endpoint (default 127.0.0.1:8080), but upstream warns it implements only basic security checks and is not recommended for production.

開啟指南

03 / 模型範圍

代表性開放權重模型

Community REAP build

Qwen3.8 Flash Next REAP-288

176BREAP-288 · MLX 4-bit · PLE on NVMe

估算記憶體: ~40.57 GB
上下文: 7.6K recommended on measured 48 GB Mac

Llama 3.1 70B

70BGGUF Q4

估算記憶體: ~46 GB
上下文: 128K

Alibaba Qwen

Qwen3.6 35B-A3B

35B3B active · Q4 / NVFP4 MoE

估算記憶體: ~25 GB
上下文: 262K native

Alibaba Qwen

Qwen3 32B

32BGGUF Q4

估算記憶體: ~24.5 GB
上下文: 32K+

Alibaba Qwen

Qwen3 30B-A3B

30B3B active · GGUF Q4 MoE

估算記憶體: ~22.5 GB
上下文: 32K+

Alibaba Qwen

Qwen3.6 27B

27BQ4 / NVFP4 estimate

估算記憶體: ~18.5 GB
上下文: 262K native

注意

適配以估算的總模型記憶體為準。驅動支援、KV 快取、多模態投影層、併發以及桌面應用都可能進一步壓縮可用餘量。目前適配：Qwen3.8 Flash Next REAP-288 (勉強適配), Llama 3.1 70B (勉強適配), Qwen3.6 35B-A3B (餘量充足), Qwen3 32B (餘量充足), Qwen3 30B-A3B (餘量充足), Qwen3.6 27B (餘量充足)。