LLM EXPLORER 61,652 MODELS INDEXED

Qwen3.8 Flash Next NVFP4 by local-inference-lab

By local-inference-lab · 20230 downloads
HF Repo ↗ Copy URL

Qwen3.8 Flash Next NVFP4 is an open-source language model by local-inference-lab. Features: 92.7b LLM, VRAM: 105GB, License: other, LLM Explorer Score: 0.37.

  8-bit Base model:quantized:qwen/qwen... Base model:qwen/qwen3.8-flash-...   Conversational   Distillation   Endpoints compatible   Image-text-to-text   Modelopt   Multimodal   Mxfp8   Nvfp4 Quantization-aware-distillatio...   Quantized   Qwen4 exp   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/local-inference-lab/Qwen3.8-Flash-Next-NVFP4 

Qwen3.8 Flash Next NVFP4 Parameters and Internals

LLM NameQwen3.8 Flash Next NVFP4
Repository 🤗https://huggingface.co/local-inference-lab/Qwen3.8-Flash-Next-NVFP4 
Base Model(s)  Qwen/Qwen3.8-Flash-Next   Qwen/Qwen3.8-Flash-Next
Model Size92.7b
Required VRAM105 GB
Updated2026-09-19
Maintainerlocal-inference-lab
Model Typeqwen4_exp
Model Files  4.2 GB: 1-of-36   1.8 GB: 2-of-36   1.7 GB: 3-of-36   4.2 GB: 4-of-36   1.8 GB: 5-of-36   4.2 GB: 6-of-36   1.8 GB: 7-of-36   4.2 GB: 8-of-36   1.8 GB: 9-of-36   4.2 GB: 10-of-36   1.8 GB: 11-of-36   4.2 GB: 12-of-36   1.8 GB: 13-of-36   4.2 GB: 14-of-36   1.8 GB: 15-of-36   4.2 GB: 16-of-36   1.8 GB: 17-of-36   4.2 GB: 18-of-36   1.8 GB: 19-of-36   4.2 GB: 20-of-36   1.8 GB: 21-of-36   4.2 GB: 22-of-36   1.8 GB: 23-of-36   4.2 GB: 24-of-36   1.8 GB: 25-of-36   4.2 GB: 26-of-36   1.8 GB: 27-of-36   4.2 GB: 28-of-36   1.8 GB: 29-of-36   4.2 GB: 30-of-36   1.8 GB: 31-of-36   4.2 GB: 32-of-36   1.8 GB: 33-of-36   4.5 GB: 34-of-36   2.8 GB: 35-of-36   0.0 GB: 36-of-36
Model ArchitectureQwen4ExpForConditionalGeneration|Qwen3_8FlashNextForConditionalGeneration
Licenseother
Model Max Length262144
Transformers Version5.12.1
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Errorsreplace