What are the hardware requirements for Qwen3.5 4B OQ4e MTP MLX?

Qwen3.5 4B OQ4e MTP MLX requires approximately 3.3 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Qwen3.5 4B OQ4e MTP MLX and how large is it?

Qwen3.5 4B OQ4e MTP MLX is developed by treeish, a model with 4b 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 Qwen3.5 4B OQ4e MTP MLX?

Qwen3.5 4B OQ4e MTP MLX is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Qwen3.5 4B OQ4e MTP MLX by treeish — VRAM 3.3GB

Name: Qwen3.5 4B OQ4e MTP MLX
Author: treeish

Qwen3.5 4B OQ4e MTP MLX is an open-source language model by treeish. Features: 4b LLM, VRAM: 3.3GB, License: apache-2.0, LLM Explorer Score: 0.24.

4-bit Base model:quantized:qwen/qwen... Base model:qwen/qwen3.5-4b Conversational Image-text-to-text Imatrix Mlx Mtp Oq Oqe Quantized Qwen3 5 Region:us Safetensors Sprig Treeish

Model Card on HF 🤗: https://huggingface.co/treeish/Qwen3.5-4B-oQ4e-MTP-MLX

Qwen3.5 4B OQ4e MTP MLX Parameters and Internals

LLM Name	Qwen3.5 4B OQ4e MTP MLX
Repository 🤗	https://huggingface.co/treeish/Qwen3.5-4B-oQ4e-MTP-MLX
Base Model(s)	Qwen/Qwen3.5-4B Qwen/Qwen3.5-4B
Model Size	4b
Required VRAM	3.3 GB
Updated	2026-08-17
Maintainer	treeish
Model Type	qwen3_5
Model Files	3.3 GB
Model Architecture	Qwen3_5ForConditionalGeneration
License	apache-2.0
Model Max Length	262144
Transformers Version	4.57.0.dev0
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Errors	replace

Rank the Qwen3.5 4B OQ4e MTP MLX 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

Quantized Models of the Qwen3.5 4B OQ4e MTP MLX

Model	Likes	Downloads	VRAM
Qwen3.5 4B OQ4e FP16 MTP MLX	1	42	3 GB

Best Alternatives to Qwen3.5 4B OQ4e MTP MLX

Best Alternatives	Context / RAM	Downloads	Likes
AgentHorse 4B	256K / 8.4 GB	878	7
Qwen3.5 4B	0K / 9.3 GB	6575502	801
LightOnOCR 3 4B	0K / 9.1 GB	2395	71
NuExtract3	0K / 0.2 GB	441570	324
Agents A1 4B	0K / 9.1 GB	177578	72
FrogNano 4B 2609	0K / 9.3 GB	733	113
Intern Decision 4B	0K / 0.1 GB	338	48
Mizan V3 3B	0K / 0.1 GB	46	14
Tev1 4B Experimental	0K / 9.3 GB	1539	16
Qwen3.5 4B Base	0K / 9.3 GB	227814	79

Note: green Score (e.g. "73.2") means that the model is better than treeish/Qwen3.5-4B-oQ4e-MTP-MLX.