What are the hardware requirements for Kimi VL A3B Thinking 2506 6bit?

Kimi VL A3B Thinking 2506 6bit requires approximately 13.9 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Kimi VL A3B Thinking 2506 6bit and how large is it?

Kimi VL A3B Thinking 2506 6bit is developed by mlx-community. 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 Kimi VL A3B Thinking 2506 6bit?

Kimi VL A3B Thinking 2506 6bit is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Kimi VL A3B Thinking 2506 6bit by mlx-community — VRAM 13.9GB

Name: Kimi VL A3B Thinking 2506 6bit
Author: mlx-community

Kimi VL A3B Thinking 2506 6bit is an open-source language model by mlx-community. Features: LLM, VRAM: 13.9GB, License: mit, Quantized, Instruction-Based, LLM Explorer Score: 0.2.

6-bit 6bit Base model:moonshotai/kimi-vl-... Base model:quantized:moonshota... Conversational Custom code Feature-extraction Image-text-to-text Instruct Kimi vl Mlx Quantized Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/mlx-community/Kimi-VL-A3B-Thinking-2506-6bit

Kimi VL A3B Thinking 2506 6bit Parameters and Internals

LLM Name	Kimi VL A3B Thinking 2506 6bit
Repository 🤗	https://huggingface.co/mlx-community/Kimi-VL-A3B-Thinking-2506-6bit
Base Model(s)	Kimi VL A3B Instruct moonshotai/Kimi-VL-A3B-Instruct
Required VRAM	13.9 GB
Updated	2026-08-08
Maintainer	mlx-community
Model Type	kimi_vl
Instruction-Based	Yes
Model Files	5.3 GB: 1-of-3 5.4 GB: 2-of-3 3.2 GB: 3-of-3
Quantization Type	6bit
Model Architecture	KimiVLForConditionalGeneration
License	mit
Model Max Length	1048576
Transformers Version	4.48.2
Padding Token	[PAD]
Vocabulary Size	163840

Rank the Kimi VL A3B Thinking 2506 6bit 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 Kimi VL A3B Thinking 2506 6bit

Best Alternatives	Context / RAM	Downloads	Likes
Kimi VL A3B Thinking 2506 8bit	0K / 17.8 GB	403	2

Note: green Score (e.g. "73.2") means that the model is better than mlx-community/Kimi-VL-A3B-Thinking-2506-6bit.