LLM EXPLORER 59,962 MODELS INDEXED

Qwen3.8 Flash Next REAP 288 MLX 4bit by sh0wie

By sh0wie · 4268 downloads
HF Repo ↗ Copy URL

Qwen3.8 Flash Next REAP 288 MLX 4bit is an open-source language model by sh0wie. Features: 21.2b LLM, VRAM: 36.2GB, License: other, Quantized, LLM Explorer Score: 0.41.

  4-bit   4bit Base model:quantized:qwen/qwen... Base model:qwen/qwen3.8-flash-...   Conversational   Image-text-to-text   Mlx   Mlx-vlm   Moe   Pruning   Quantized   Qwen   Qwen4 exp   Reap   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/sh0wie/Qwen3.8-Flash-Next-REAP-288-MLX-4bit 

Qwen3.8 Flash Next REAP 288 MLX 4bit Parameters and Internals

LLM NameQwen3.8 Flash Next REAP 288 MLX 4bit
Repository 🤗https://huggingface.co/sh0wie/Qwen3.8-Flash-Next-REAP-288-MLX-4bit 
Base Model(s)  Qwen/Qwen3.8-Flash-Next   Qwen/Qwen3.8-Flash-Next
Model Size21.2b
Required VRAM36.2 GB
Updated2026-08-30
Maintainersh0wie
Model Typeqwen4_exp
Model Files  0.5 GB: 1-of-131   0.5 GB: 2-of-131   0.3 GB: 3-of-131   0.5 GB: 4-of-131   0.8 GB: 5-of-131   1.0 GB: 6-of-131   1.0 GB: 7-of-131   1.0 GB: 8-of-131   1.0 GB: 9-of-131   1.0 GB: 10-of-131   1.0 GB: 11-of-131   1.0 GB: 12-of-131   1.0 GB: 13-of-131   1.0 GB: 14-of-131   1.0 GB: 15-of-131   1.0 GB: 16-of-131   1.0 GB: 17-of-131   1.0 GB: 18-of-131   1.0 GB: 19-of-131   1.0 GB: 20-of-131   1.0 GB: 21-of-131   1.0 GB: 22-of-131   1.0 GB: 23-of-131   1.0 GB: 24-of-131   1.0 GB: 25-of-131   1.0 GB: 26-of-131   1.0 GB: 27-of-131   1.0 GB: 28-of-131   1.0 GB: 29-of-131   1.0 GB: 30-of-131   1.0 GB: 31-of-131   1.0 GB: 32-of-131   1.0 GB: 33-of-131   1.0 GB: 34-of-131   1.0 GB: 35-of-131   1.0 GB: 36-of-131   0.5 GB: 37-of-131   0.5 GB: 38-of-131   0.3 GB: 39-of-131   0.5 GB: 40-of-131   0.3 GB: 41-of-131   0.5 GB: 42-of-131
Quantization Type4bit
Model ArchitectureQwen4ExpForConditionalGeneration
Licenseother
Transformers Version5.8.0.dev0