What are the hardware requirements for Swift Qwen3.8 27B PARO MXFP6?

Swift Qwen3.8 27B PARO MXFP6 requires approximately 25.1 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Swift Qwen3.8 27B PARO MXFP6 and how large is it?

Swift Qwen3.8 27B PARO MXFP6 is developed by hugypufy, 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 Swift Qwen3.8 27B PARO MXFP6?

Swift Qwen3.8 27B PARO MXFP6 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Swift Qwen3.8 27B PARO MXFP6 by hugypufy — VRAM 25.1GB

Name: Swift Qwen3.8 27B PARO MXFP6
Author: hugypufy

Swift Qwen3.8 27B PARO MXFP6 is an open-source language model by hugypufy. Features: 27b LLM, VRAM: 25.1GB, License: other, LLM Explorer Score: 0.31.

6-bit Base model:quantized:ukisai/sw... Base model:ukisai/swift-qwen3.... Conversational Endpoints compatible Image-text-to-text Mxfp6 Paroquant Paroquant mxfp6 Quantized Qwen3 5 Rdna4 Region:us Rocm Safetensors Sharded Tensorflow Vllm W6a8

Model Card on HF 🤗: https://huggingface.co/hugypufy/Swift-Qwen3.8-27B-PARO-MXFP6

Swift Qwen3.8 27B PARO MXFP6 Parameters and Internals

LLM Name	Swift Qwen3.8 27B PARO MXFP6
Repository 🤗	https://huggingface.co/hugypufy/Swift-Qwen3.8-27B-PARO-MXFP6
Base Model(s)	ukisai/Swift-Qwen3.8-27b ukisai/Swift-Qwen3.8-27b
Model Size	27b
Required VRAM	25.1 GB
Updated	2026-09-19
Maintainer	hugypufy
Model Type	qwen3_5
Model Files	2.1 GB: 1-of-18 1.2 GB: 2-of-18 2.5 GB: 3-of-18 1.6 GB: 4-of-18 0.8 GB: 5-of-18 1.6 GB: 6-of-18 0.8 GB: 7-of-18 1.6 GB: 8-of-18 0.8 GB: 9-of-18 1.6 GB: 10-of-18 0.8 GB: 11-of-18 1.6 GB: 12-of-18 0.8 GB: 13-of-18 1.6 GB: 14-of-18 0.8 GB: 15-of-18 1.6 GB: 16-of-18 0.8 GB: 17-of-18 2.5 GB: 18-of-18
Model Architecture	Qwen3_5ForConditionalGeneration
License	other
Model Max Length	262144
Transformers Version	5.8.0.dev0
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Errors	replace

Rank the Swift Qwen3.8 27B PARO MXFP6 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 Swift Qwen3.8 27B PARO MXFP6

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	145	1
Qwen3.6 27B	0K / 55.4 GB	6728138	2197
Qwen3.8 27B	0K / 55.6 GB	2	8753
Qwen3.8 27B FP8	0K / 0.4 GB	123157	416
Qwen3.8 27B MLX	0K / 16 GB	1411	18
Qwen3.8 27B	0K / 55.6 GB	2727	21
Qwen3.6 27B NVFP4	0K / 22 GB	1887525	419
Qwen3.6 27B FP8	0K / 0.4 GB	7740472	342
Qwen3.5 27B	0K / 55.2 GB	2536294	1028

Note: green Score (e.g. "73.2") means that the model is better than hugypufy/Swift-Qwen3.8-27B-PARO-MXFP6.