LLM EXPLORER 57,252 MODELS INDEXED

Karakuri Vl 2 8B Thinking 2603 by karakuri-ai

By karakuri-ai · 64 downloads
HF Repo ↗ Copy URL

Karakuri Vl 2 8B Thinking 2603 is an open-source language model by karakuri-ai. Features: 8b LLM, VRAM: 17.5GB, License: apache-2.0, LLM Explorer Score: 0.24.

Base model:finetune:qwen/qwen3... Base model:qwen/qwen3-vl-8b-th...   Conversational   En   Endpoints compatible   Image-text-to-text   Ja   Qwen3 vl   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/karakuri-ai/karakuri-vl-2-8b-thinking-2603 

Karakuri Vl 2 8B Thinking 2603 Parameters and Internals

LLM NameKarakuri Vl 2 8B Thinking 2603
Repository 🤗https://huggingface.co/karakuri-ai/karakuri-vl-2-8b-thinking-2603 
Base Model(s)  Qwen/Qwen3-VL-8B-Thinking   Qwen/Qwen3-VL-8B-Thinking
Model Size8b
Required VRAM17.5 GB
Updated2026-08-08
Maintainerkarakuri-ai
Model Typeqwen3_vl
Model Files  5.0 GB: 1-of-4   4.9 GB: 2-of-4   4.9 GB: 3-of-4   2.7 GB: 4-of-4
Supported Languagesja en
Model ArchitectureQwen3VLForConditionalGeneration
Licenseapache-2.0
Model Max Length262144
Transformers Version4.57.0.dev0
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Errorsreplace

Quantized Models of the Karakuri Vl 2 8B Thinking 2603

Model
Likes
Downloads
VRAM
...Vl 2 8B Thinking 2603 MLX 8bit179 GB

Best Alternatives to Karakuri Vl 2 8B Thinking 2603

Best Alternatives
Context / RAM
Downloads
Likes
Qwen3 VL 8B Instruct0K / 17.5 GB44943291037
Holo2 8B0K / 17.5 GB192430
Qwen3 VL 8B Instruct FP80K / 10.6 GB310860477
SpatialCLI 8B0K / 17.5 GB914
UI Venus 1.5 8B0K / 17.5 GB843529
MAI UI 8B0K / 17.6 GB2224199
... Simple Path Dense S100 Step900K / 17.5 GB131
ZwZ 8B0K / 17.6 GB258648
EvoCUA 8B 202601050K / 17.5 GB441716
Qwen3 VL 8B Instruct Bf160K / 17.6 GB1273
Note: green Score (e.g. "73.2") means that the model is better than karakuri-ai/karakuri-vl-2-8b-thinking-2603.