LLM EXPLORER 61,405 MODELS INDEXED

Qwen3.8 Flash Next MLX Serve IQ MLX 3.3bpw by ddalcu

By ddalcu · 604 downloads
HF Repo ↗ Copy URL

Qwen3.8 Flash Next MLX Serve IQ MLX 3.3bpw is an open-source language model by ddalcu. Features: 98.4b LLM, VRAM: 0.1GB, License: other, LLM Explorer Score: 0.35.

  4-bit Base model:quantized:qwen/qwen... Base model:qwen/qwen3.8-flash-...   Conversational   Imatrix   Mlx   Mlx-serve   Moe   Ngram-embedding   Qwen4 exp   Region:us   Safetensors   Sparse-attention
Model Card on HF 🤗: https://huggingface.co/ddalcu/Qwen3.8-Flash-Next-MLX-Serve-iQ-MLX-3.3bpw 

Qwen3.8 Flash Next MLX Serve IQ MLX 3.3bpw Parameters and Internals

LLM NameQwen3.8 Flash Next MLX Serve IQ MLX 3.3bpw
Repository 🤗https://huggingface.co/ddalcu/Qwen3.8-Flash-Next-MLX-Serve-iQ-MLX-3.3bpw 
Base Model(s)  Qwen/Qwen3.8-Flash-Next   Qwen/Qwen3.8-Flash-Next
Model Size98.4b
Required VRAM0.1 GB
Updated2026-09-16
Maintainerddalcu
Model Typeqwen4_exp
Model Files  0.1 GB   0.7 GB   0.3 GB   0.7 GB   0.3 GB   0.1 GB   0.7 GB   0.3 GB   0.7 GB   0.4 GB   0.7 GB   0.3 GB   0.7 GB   0.3 GB   0.7 GB   0.3 GB   0.5 GB   0.4 GB   0.7 GB   0.3 GB   0.7 GB   0.4 GB   0.7 GB   0.3 GB   0.7 GB   0.5 GB   0.7 GB   0.3 GB   1.0 GB   0.7 GB   0.9 GB   0.7 GB   0.4 GB   0.7 GB   0.4 GB   0.7 GB   0.5 GB   0.7 GB   0.4 GB   0.7 GB   0.3 GB   0.7 GB   0.4 GB
Model ArchitectureQwen4ExpForConditionalGeneration
Licenseother
Transformers Version5.8.0.dev0