What are the hardware requirements for Swift Qwen3.8 27B W4A16 AutoRound?

Swift Qwen3.8 27B W4A16 AutoRound requires approximately 9 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Swift Qwen3.8 27B W4A16 AutoRound and how large is it?

Swift Qwen3.8 27B W4A16 AutoRound is developed by jamesbrunet, 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 W4A16 AutoRound?

Swift Qwen3.8 27B W4A16 AutoRound 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 W4A16 AutoRound by jamesbrunet — VRAM 9GB

Name: Swift Qwen3.8 27B W4A16 AutoRound
Author: jamesbrunet

Swift Qwen3.8 27B W4A16 AutoRound is an open-source language model by jamesbrunet. Features: 27b LLM, VRAM: 9GB, License: other, LLM Explorer Score: 0.34.

Auto-round Base model:quantized:ukisai/sw... Base model:ukisai/swift-qwen3.... Compressed-tensors Conversational Endpoints compatible Image-text-to-text Int4 Qwen3 Qwen3 5 Region:us Safetensors Sharded Tensorflow Vllm W4a16

Model Card on HF 🤗: https://huggingface.co/jamesbrunet/Swift-Qwen3.8-27b-W4A16-AutoRound

Swift Qwen3.8 27B W4A16 AutoRound Parameters and Internals

LLM Name	Swift Qwen3.8 27B W4A16 AutoRound
Repository 🤗	https://huggingface.co/jamesbrunet/Swift-Qwen3.8-27b-W4A16-AutoRound
Base Model(s)	ukisai/Swift-Qwen3.8-27b ukisai/Swift-Qwen3.8-27b
Model Size	27b
Required VRAM	9 GB
Updated	2026-09-18
Maintainer	jamesbrunet
Model Type	qwen3_5
Model Files	0.2 GB: 1-of-67 0.2 GB: 2-of-67 0.2 GB: 3-of-67 0.2 GB: 4-of-67 0.2 GB: 5-of-67 0.2 GB: 6-of-67 0.2 GB: 7-of-67 0.2 GB: 8-of-67 0.2 GB: 9-of-67 0.2 GB: 10-of-67 0.2 GB: 11-of-67 0.2 GB: 12-of-67 0.2 GB: 13-of-67 0.2 GB: 14-of-67 0.2 GB: 15-of-67 0.2 GB: 16-of-67 0.2 GB: 17-of-67 0.2 GB: 18-of-67 0.2 GB: 19-of-67 0.2 GB: 20-of-67 0.2 GB: 21-of-67 0.2 GB: 22-of-67 0.2 GB: 23-of-67 0.2 GB: 24-of-67 0.2 GB: 25-of-67 0.2 GB: 26-of-67 0.2 GB: 27-of-67 0.2 GB: 28-of-67 0.2 GB: 29-of-67 0.2 GB: 30-of-67 0.2 GB: 31-of-67 0.2 GB: 32-of-67 0.2 GB: 33-of-67 0.2 GB: 34-of-67 0.2 GB: 35-of-67 0.2 GB: 36-of-67 0.2 GB: 37-of-67 0.2 GB: 38-of-67 0.2 GB: 39-of-67 0.2 GB: 40-of-67 0.2 GB: 41-of-67 0.2 GB: 42-of-67 0.2 GB: 43-of-67 0.2 GB: 44-of-67 0.2 GB: 45-of-67
Model Architecture	Qwen3_5ForConditionalGeneration
License	other
Transformers Version	5.16.1

Rank the Swift Qwen3.8 27B W4A16 AutoRound 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 W4A16 AutoRound

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	143	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 jamesbrunet/Swift-Qwen3.8-27b-W4A16-AutoRound.