LLM EXPLORER 59,962 MODELS INDEXED

Qwen3.8 Flash Next MLX OQ6 MTP by Vontra

By Vontra · 1384 downloads
HF Repo ↗ Copy URL

Qwen3.8 Flash Next MLX OQ6 MTP is an open-source language model by Vontra. Features: 45.4b LLM, VRAM: 157GB, License: other, LLM Explorer Score: 0.34.

  6-bit   Apple-silicon Base model:quantized:qwen/qwen... Base model:qwen/qwen3.8-flash-...   Conversational   Image-text-to-text   Mixture-of-experts   Mlx   Mlx-vlm   Mtp   Omlx   Oq   Quantized   Qwen   Qwen3.8   Qwen4 exp   Region:us   Safetensors   Sharded   Speculative-decoding   Tensorflow   Vision   Vision-language
Model Card on HF 🤗: https://huggingface.co/Vontra/Qwen3.8-Flash-Next-MLX-oQ6-MTP 

Qwen3.8 Flash Next MLX OQ6 MTP Parameters and Internals

LLM NameQwen3.8 Flash Next MLX OQ6 MTP
Repository 🤗https://huggingface.co/Vontra/Qwen3.8-Flash-Next-MLX-oQ6-MTP 
Base Model(s)  Qwen/Qwen3.8-Flash-Next   Qwen/Qwen3.8-Flash-Next
Model Size45.4b
Required VRAM157 GB
Updated2026-08-30
MaintainerVontra
Model Typeqwen4_exp
Model Files  5.3 GB: 1-of-31   5.2 GB: 2-of-31   5.2 GB: 3-of-31   5.2 GB: 4-of-31   5.2 GB: 5-of-31   5.2 GB: 6-of-31   5.2 GB: 7-of-31   5.2 GB: 8-of-31   5.0 GB: 9-of-31   5.3 GB: 10-of-31   5.1 GB: 11-of-31   5.1 GB: 12-of-31   5.1 GB: 13-of-31   5.1 GB: 14-of-31   5.1 GB: 15-of-31   5.1 GB: 16-of-31   5.1 GB: 17-of-31   5.1 GB: 18-of-31   5.1 GB: 19-of-31   5.1 GB: 20-of-31   5.1 GB: 21-of-31   5.1 GB: 22-of-31   5.1 GB: 23-of-31   5.1 GB: 24-of-31   5.1 GB: 25-of-31   5.1 GB: 26-of-31   5.1 GB: 27-of-31   5.1 GB: 28-of-31   5.1 GB: 29-of-31   5.1 GB: 30-of-31   3.0 GB: 31-of-31
Model ArchitectureQwen4ExpForConditionalGeneration
Licenseother
Model Max Length262144
Transformers Version5.8.0.dev0
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Errorsreplace