LLM EXPLORER 59,962 MODELS INDEXED

Qwen3.8 Flash Next FP6 by Soomin33

By Soomin33 · 761 downloads
HF Repo ↗ Copy URL

Qwen3.8 Flash Next FP6 is an open-source language model by Soomin33. Features: 98b LLM, VRAM: 4.2GB, License: other, LLM Explorer Score: 0.33.

  8-bit Base model:quantized:qwen/qwen... Base model:qwen/qwen3.8-flash-...   Conversational   Fp6   Mixture-of-experts   N-gram-embedding   Quantized   Qwen4 exp   Region:us   Safetensors   Sglang
Model Card on HF 🤗: https://huggingface.co/Soomin33/Qwen3.8-Flash-Next-FP6 

Qwen3.8 Flash Next FP6 Parameters and Internals

LLM NameQwen3.8 Flash Next FP6
Repository 🤗https://huggingface.co/Soomin33/Qwen3.8-Flash-Next-FP6 
Base Model(s)  Qwen/Qwen3.8-Flash-Next-FP8   Qwen/Qwen3.8-Flash-Next-FP8
Model Size98b
Required VRAM4.2 GB
Updated2026-08-30
MaintainerSoomin33
Model Typeqwen4_exp
Model Files  4.3 GB   4.3 GB   4.3 GB   4.3 GB   4.3 GB   4.3 GB   4.3 GB   4.3 GB   4.3 GB   4.3 GB   4.3 GB   4.3 GB   4.3 GB   4.3 GB   4.3 GB   4.3 GB   4.3 GB   4.3 GB   4.3 GB   4.3 GB   4.3 GB   4.3 GB   4.3 GB   4.3 GB   4.2 GB
Model ArchitectureQwen4ExpForConditionalGeneration
Licenseother
Model Max Length262144
Transformers Version5.8.0.dev0
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Errorsreplace