LMMs-Lab
首页文章项目关于
English
登录注册 ↗
首页文章项目关于English
登录注册 ↗

构建

通向多模态智能的道路。

多模态是好的。

我们相信,多模态是智能的必要条件。所以我们构建能看、能听、能在世界中行动,并从行动的反馈中学习的模型。

探索研究 ↗
关于实验室 ↗
LMMS-LAB // BREACH ACTIVENEURAL WEIGHT EXTRACTION
1/7LIVE
_
thinking:
_
实验室内部模型 · 评测 · 开源

精选文章

From Pixels to Actions – Towards Native Vision-Language-Action Policies
2026年11月3日

From Pixels to Actions – Towards Native Vision-Language-Action Policies

NEO-VLA feeds pixels straight into a native vision-language backbone. Control draws on its shallow layers, and native pretraining drives the gains.

models
阅读文章 ↗

最新发布

查看全部文章 ↗
LLaVA-OneVision-2: Towards Next-Generation Perceptual Intelligence
2026年4月

LLaVA-OneVision-2: Towards Next-Generation Perceptual Intelligence

models
OneVision Encoder: Codec-Aligned Sparsity as a Foundational Principle for Multimodal Intelligence
2026年1月

OneVision Encoder: Codec-Aligned Sparsity as a Foundational Principle for Multimodal Intelligence

models
LLaVA-OneVision-1.5: Fully Open Framework for Democratized Multimodal Training
2025年9月

LLaVA-OneVision-1.5: Fully Open Framework for Democratized Multimodal Training

models
LongVT: Incentivizing "Thinking with Long Videos" via Native Tool Calling
2025年11月

LongVT: Incentivizing "Thinking with Long Videos" via Native Tool Calling

models

一起做研究,会更好。

开始使用
LMMs-Lab2026 LMMs-Lab
GitHubDiscordX回到顶部 ↑