What are the hardware requirements for Ukisai Swift Qwen3.8 27B MTPLX?

Ukisai Swift Qwen3.8 27B MTPLX requires approximately 15.1 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Ukisai Swift Qwen3.8 27B MTPLX and how large is it?

Ukisai Swift Qwen3.8 27B MTPLX is developed by arsis-dev, 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 Ukisai Swift Qwen3.8 27B MTPLX?

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

Ukisai Swift Qwen3.8 27B MTPLX by arsis-dev — VRAM 15.1GB

Name: Ukisai Swift Qwen3.8 27B MTPLX
Author: arsis-dev

Ukisai Swift Qwen3.8 27B MTPLX is an open-source language model by arsis-dev. Features: 27b LLM, VRAM: 15.1GB, License: other, LLM Explorer Score: 0.28.

4-bit Apple-silicon Base model:quantized:ukisai/sw... Base model:ukisai/swift-qwen3.... Conversational Image-text-to-text Mlx Mtp Mtplx Qwen3 5 Qwen3 8 Region:us Safetensors Sharded Speculative-decoding Tensorflow

Model Card on HF 🤗: https://huggingface.co/arsis-dev/ukisai-Swift-Qwen3.8-27b-MTPLX

Ukisai Swift Qwen3.8 27B MTPLX Parameters and Internals

LLM Name	Ukisai Swift Qwen3.8 27B MTPLX
Repository 🤗	https://huggingface.co/arsis-dev/ukisai-Swift-Qwen3.8-27b-MTPLX
Base Model(s)	ukisai/Swift-Qwen3.8-27b ukisai/Swift-Qwen3.8-27b
Model Size	27b
Required VRAM	15.1 GB
Updated	2026-09-20
Maintainer	arsis-dev
Model Type	qwen3_5
Model Files	5.3 GB: 1-of-3 5.3 GB: 2-of-3 4.5 GB: 3-of-3 0.9 GB 0.8 GB
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 Ukisai Swift Qwen3.8 27B MTPLX 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 Ukisai Swift Qwen3.8 27B MTPLX

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	86	3
Qwen3.8 27B NPU2	256K / GB	252	4
JEV 27B VL	0K / 55.6 GB	1525286	966
Clef	0K / 54.7 GB	22043	2237
Arete 27B	0K / 54.7 GB	2407	346
Qwen3.6 27B	0K / 55.4 GB	6728138	2197
Qwen3.8 27B	0K / 55.6 GB	2	8753
Qwen3.8 27B NVFP4 DGX SPARK V2	0K / 19.8 GB	11317	113
MiniMax H3 Image PE	0K / 18 GB	600	52

Note: green Score (e.g. "73.2") means that the model is better than arsis-dev/ukisai-Swift-Qwen3.8-27b-MTPLX.