What are the hardware requirements for Qwen3.6 35B A3B Mirage Mxfp4 Mlx?

Qwen3.6 35B A3B Mirage Mxfp4 Mlx requires approximately 19.3 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Qwen3.6 35B A3B Mirage Mxfp4 Mlx and how large is it?

Qwen3.6 35B A3B Mirage Mxfp4 Mlx is developed by nightmedia, a model with 35b 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 Qwen3.6 35B A3B Mirage Mxfp4 Mlx?

Qwen3.6 35B A3B Mirage Mxfp4 Mlx is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Qwen3.6 35B A3B Mirage Mxfp4 Mlx by nightmedia — VRAM 19.3GB

Name: Qwen3.6 35B A3B Mirage Mxfp4 Mlx
Author: nightmedia

Qwen3.6 35B A3B Mirage Mxfp4 Mlx is an open-source language model by nightmedia. Features: 35b LLM, VRAM: 19.3GB, License: apache-2.0, Instruction-Based, LLM Explorer Score: 0.31.

4-bit Base model:adapter:allspark-re... Base model:allspark-research/i... Chain-of-thought Coding Conversational Deckard(qx) Distillation En Endpoints compatible Es Experimental Fable Humour Image-text-to-text Instruct Instruction-tuned Ja Long-cot Lora Math Merge Mergekit Mlx Multilingual Qwen Qwen3.5 Qwen3.6 Qwen3 5 Qwen3 5 moe Qwen3 6 Reasoning Region:us Research Safetensors Sft Sharded Stem Tensorflow Unsloth Zh

Model Card on HF 🤗: https://huggingface.co/nightmedia/Qwen3.6-35B-A3B-Mirage-mxfp4-mlx

Qwen3.6 35B A3B Mirage Mxfp4 Mlx Parameters and Internals

LLM Name	Qwen3.6 35B A3B Mirage Mxfp4 Mlx
Repository 🤗	https://huggingface.co/nightmedia/Qwen3.6-35B-A3B-Mirage-mxfp4-mlx
Base Model(s)	shunzou05/TerminalHorizon-35B-A3B Iris Mini thomsonreuters/Thomson-1.0-Small nightmedia/Qwen3.6-35B-A3B-Fable-Holo3.1 nightmedia/Qwen3.6-35B-A3B-FSM Qwen AgentWorld 35B A3B orcarouter/Nex-N2.5-mini-Uncensored ...AgentWorld 35B A3B Abliterated shunzou05/TerminalHorizon-35B-A3B AllSpark-Research/Iris-mini thomsonreuters/Thomson-1.0-Small nightmedia/Qwen3.6-35B-A3B-Fable-Holo3.1 nightmedia/Qwen3.6-35B-A3B-FSM Qwen/Qwen-AgentWorld-35B-A3B orcarouter/Nex-N2.5-mini-Uncensored Jiunsong/SuperQwen-AgentWorld-35B-A3B-abliterated
Model Size	35b
Required VRAM	19.3 GB
Updated	2026-10-22
Maintainer	nightmedia
Model Type	qwen3_5_moe
Instruction-Based	Yes
Model Files	5.3 GB: 1-of-4 5.4 GB: 2-of-4 5.4 GB: 3-of-4 3.2 GB: 4-of-4
Supported Languages	en zh ja es
Model Architecture	Qwen3_5MoeForConditionalGeneration
License	apache-2.0
Model Max Length	262144
Transformers Version	5.17.0
Tokenizer Class	TokenizersBackend
Padding Token	<\|endoftext\|>
Errors	replace

Rank the Qwen3.6 35B A3B Mirage Mxfp4 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

Best Alternatives to Qwen3.6 35B A3B Mirage Mxfp4 Mlx

Best Alternatives	Context / RAM	Downloads	Likes
Keural Nova 1.0	0K / 70.9 GB	22	1
Keural Nova 1.1	0K / 70.9 GB	18	1
Qwen3.6 35B A3B Brainwaves Nex	0K / 72 GB	13	0
Tema Q X5 Thinking	0K / 70.2 GB	21	1
... A3B Brainwaves Nex Uncensored	0K / 72 GB	242	1
...ves Nex Uncensored Qx64 Hi Mlx	0K / 30.8 GB	258	1