LLM EXPLORER 58,608 MODELS INDEXED

Kimi VL A3B Thinking 2506 8bit by mlx-community

By mlx-community · 403 downloads
HF Repo ↗ Copy URL

Kimi VL A3B Thinking 2506 8bit is an open-source language model by mlx-community. Features: LLM, VRAM: 17.8GB, License: mit, Quantized, Instruction-Based, LLM Explorer Score: 0.2.

  8-bit   8bit 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-2506-8bit 

Kimi VL A3B Thinking 2506 8bit Parameters and Internals

LLM NameKimi VL A3B Thinking 2506 8bit
Repository 🤗https://huggingface.co/mlx-community/Kimi-VL-A3B-Thinking-2506-8bit 
Base Model(s)  Kimi VL A3B Instruct   moonshotai/Kimi-VL-A3B-Instruct
Required VRAM17.8 GB
Updated2026-08-08
Maintainermlx-community
Model Typekimi_vl
Instruction-BasedYes
Model Files  5.2 GB: 1-of-4   5.4 GB: 2-of-4   5.2 GB: 3-of-4   2.0 GB: 4-of-4
Quantization Type8bit
Model ArchitectureKimiVLForConditionalGeneration
Licensemit
Model Max Length1048576
Transformers Version4.48.2
Padding Token[PAD]
Vocabulary Size163840

Best Alternatives to Kimi VL A3B Thinking 2506 8bit

Best Alternatives
Context / RAM
Downloads
Likes
Kimi VL A3B Thinking 2506 6bit0K / 13.9 GB3852