What are the hardware requirements for Qwen3.8 27B Mixed NVFP4 FP8?

Qwen3.8 27B Mixed NVFP4 FP8 requires approximately 3.7 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Qwen3.8 27B Mixed NVFP4 FP8 and how large is it?

Qwen3.8 27B Mixed NVFP4 FP8 is developed by primitive-ai, a model with 27b 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 27B Mixed NVFP4 FP8?

Qwen3.8 27B Mixed NVFP4 FP8 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Qwen3.8 27B Mixed NVFP4 FP8 by primitive-ai — VRAM 3.7GB

Name: Qwen3.8 27B Mixed NVFP4 FP8
Author: primitive-ai

Qwen3.8 27B Mixed NVFP4 FP8 is an open-source language model by primitive-ai. Features: 27b LLM, VRAM: 3.7GB, License: apache-2.0, LLM Explorer Score: 0.26.

8-bit A100 Base model:quantized:qwen/qwen... Base model:qwen/qwen3.8-27b Blackwell Compressed-tensors Conversational Endpoints compatible Fp8 Image-text-to-text Mixed-precision Nvfp4 Quantized Qwen3 5 Region:us Safetensors Speculative-decoding Vllm

Model Card on HF 🤗: https://huggingface.co/primitive-ai/Qwen3.8-27B-mixed-NVFP4-FP8

Qwen3.8 27B Mixed NVFP4 FP8 Parameters and Internals

LLM Name	Qwen3.8 27B Mixed NVFP4 FP8
Repository 🤗	https://huggingface.co/primitive-ai/Qwen3.8-27B-mixed-NVFP4-FP8
Base Model(s)	Qwen/Qwen3.8-27B Qwen/Qwen3.8-27B
Model Size	27b
Required VRAM	3.7 GB
Updated	2026-08-19
Maintainer	primitive-ai
Model Type	qwen3_5
Model Files	5.6 GB 4.3 GB 4.3 GB 4.3 GB 3.7 GB
Model Architecture	Qwen3_5ForConditionalGeneration
License	apache-2.0
Model Max Length	262144
Transformers Version	5.8.0.dev0
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Errors	replace

Rank the Qwen3.8 27B Mixed NVFP4 FP8 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 27B Mixed NVFP4 FP8

Best Alternatives	Context / RAM	Downloads	Likes
...wopus3.8 27B Flash Mlx OQ4e 1M	1024K / 16.9 GB	211	1
Agens Pilot	1024K / 54.7 GB	93	3
Qwen3.8 27B NPU2	256K / GB	252	4
JEV 27B VL	0K / 55.6 GB	1525286	966
Clef	0K / 54.7 GB	21094	2217
Qwen3.6 27B	0K / 55.4 GB	6728138	2197
Qwen3.8 27B	0K / 55.6 GB	2	8753
Qwen3.8 27B	0K / 55.6 GB	2727	21
Qwen3.8 27B MLX	0K / 16 GB	1411	18
Qwen3.6 27B NVFP4	0K / 22 GB	1887525	419

Note: green Score (e.g. "73.2") means that the model is better than primitive-ai/Qwen3.8-27B-mixed-NVFP4-FP8.