PulseAugur
EN
LIVE 16:36:59
中文(ZH) 小鹏正式发布TuringViT高效视觉编码器

XPeng launches TuringViT vision encoder for AI applications

XPeng has officially launched its TuringViT efficient vision encoder, designed to systematically reconstruct the architecture, data paradigm, and training process for vision encoders in the VLM/VLA era. This new encoder will fully support XPeng's intelligent driving, intelligent cockpit, and IRON humanoid robot business scenarios. The announcement was made by 36Kr, which also noted unrelated news about Shenglian Biology receiving approval for a swine pseudorabies vaccine and a preview of WAIC 2026 trends, alongside a mention of OpenAI halting GPT-6. AI

IMPACT XPeng's new vision encoder aims to improve AI capabilities in autonomous driving, cockpits, and robotics.

RANK_REASON The item announces a new product/encoder from a specific company, not a frontier model release or significant industry-wide event.

Read on 36氪 (36Kr) →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

XPeng launches TuringViT vision encoder for AI applications

COVERAGE [1]

  1. 36氪 (36Kr) TIER_1 中文(ZH) ·

    Xpeng Officially Releases TuringViT Efficient Vision Encoder

    36氪获悉,近日,小鹏正式发布TuringViT高效视觉编码器,面向VLM/VLA时代系统性重构视觉编码器的架构设计、数据范式与训练流程。TuringViT将全面支撑智能驾驶、智能座舱、IRON人形机器人三大业务场景。