What are the hardware requirements for Qwen3.8 Flash Next NVFP4?

Qwen3.8 Flash Next NVFP4 requires approximately 105 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Qwen3.8 Flash Next NVFP4 and how large is it?

Qwen3.8 Flash Next NVFP4 is developed by local-inference-lab, a model with 92.7b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Qwen3.8 Flash Next NVFP4?

Qwen3.8 Flash Next NVFP4 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Qwen3.8 Flash Next NVFP4 by local-inference-lab — VRAM 105GB

Name: Qwen3.8 Flash Next NVFP4
Author: local-inference-lab

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: 0.33.

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 Name	Qwen3.8 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 Size	92.7b
Required VRAM	105 GB
Updated	2026-09-19
Maintainer	local-inference-lab
Model Type	qwen4_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 Architecture	Qwen4ExpForConditionalGeneration\|Qwen3_8FlashNextForConditionalGeneration
License	other
Model Max Length	262144
Transformers Version	5.12.1
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Errors	replace

Rank the Qwen3.8 Flash Next NVFP4 capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to Qwen3.8 Flash Next NVFP4

Best Alternatives	Context / RAM	Downloads	Likes
...8 Flash Next NVFP4 Abliterated	0K / 105 GB	420	2