LocalForge
Torna alla guida
MODEL / COMMUNITY REAP BUILD

Qwen3.8 Flash Next REAP-288

Model card checked 2026-09-16 · Memory fit estimated; speed unknown unless a separately linked measurement matches your configuration.

Also searched as: qwen3.8-flash-next-reap-288-mlx-4bit

Community-pruned conversion, not the stock checkpoint. The 39 GB model / ~40.57 GB total resident path requires oMLX 0.6.4+ and streams only the sparse PLE table from NVMe. A published M4 Pro 48 GB test measured 26–29 tok/s and recommended 7,680 tokens; do not generalize it to arbitrary models or Macs.

Fonte primaria
Parametri totali176B
Memoria stimata~40.57 GBREAP-288 · MLX 4-bit · PLE on NVMe
Contesto7.6K recommended on measured 48 GB MacText
LicenseQwen CommunityMultilingual; post-pruning quality not independently established
01 / IDEALE PER
  • 48 GB Mac experiments
  • coding agents
  • model-specific NVMe streaming
02 / COMPATIBILITÀ STIMATA
Mac Studio M5 Ultra · 96 GBComodoMacBook Pro M5 Max · 64 GBAl limiteMac Studio M3 Ultra · 96 GBComodoMac Studio M3 Ultra · 256 GBComodoApple silicon · 64 GBAl limite
03 / ESEGUI CON

These are runtime examples, not validated installation recipes. Fit estimates assume quantized files; official repositories may hold full-precision weights. MLX, GGUF and TensorRT require matching conversions or engines.

oMLXAdvancedomlx serve --model-dir ~/models --memory-guard-gb 48
Attenzione

Community-pruned conversion, not the stock checkpoint. The 39 GB model / ~40.57 GB total resident path requires oMLX 0.6.4+ and streams only the sparse PLE table from NVMe. A published M4 Pro 48 GB test measured 26–29 tok/s and recommended 7,680 tokens; do not generalize it to arbitrary models or Macs.