Mini toolsPenguin VL 2B
By Tencent
Penguin-VL pairs Qwen3 backbones with an LLM-based vision encoder approach (instead of contrastive-pretrained encoders) to preserve fine-grained visual cues for dense captioning and complex VLM reasoning, released as a 2B-scale variant with inference code and demos.
Multimodal
Overview
Penguin-VL-2B is a compact vision-language model that uses an LLM-based vision encoder to push efficiency limits in multimodal reasoning.
About Tencent
Tencent is a world-leading internet and technology company that develops innovative products and services to improve the quality of life of people around the world.
Industry:
Technology, Information and Media
Company Size:
110558
Location:
Shenzhen, CN
Website:
tencent.com
Last updated: April 6, 2026
