帖子、文章和讨论

Community Articles

Granite 4.0 Nano: Just how small can you go?

and 1 other •

Why Did MiniMax M2 End Up as a Full Attention Model?

On the Shifting Global Compute Landscape

and 1 other •

The World’s First and Best Speed Painting Software

🛡️ Nemotron PII: Synthesized Data for Privacy-Preserving AI

What makes good reasoning data

NVIDIA Isaac GR00T in LeRobot

and 4 others •

Aligning to What? Rethinking Agent Generalization in MiniMax M2

Cosmos Predict 2.5 & Transfer 2.5: Evolving the World Foundation Models for Physical AI

NVIDIA Releases 8 Million Sample Open Dataset and Tooling for OCR, Image Reasoning, Image and Video QA Tasks

and 6 others •

How to Build a Healthcare Robot from Simulation to Deployment with NVIDIA Isaac for Healthcare

Can Your LLM Think Like a Professional? Introducing ProfBench

and 7 others •

3+ Years of ML & Society at Hugging Face 🤗🤝🧑‍🤝‍🧑

and 3 others •

LightOnOCR-1B: The Case for End-to-End and Efficient Domain-Specific Vision-Language Models for OCR

and 2 others •

Code a simple RAG from scratch

Nemotron-Personas-USA: Synthesized Data for Sovereign AI

and 2 others •

The 1 Billion Token Challenge: Finding the Perfect Pre-training Mix

about 23 hours ago

KV Caching Explained: Optimizing Transformer Inference Efficiency

svara-TTS — Open Multilingual TTS for India’s Voices

Let's talk about LLM evaluation

lerobotrobotics

LeRobot v0.4.0：全面提升开源机器人的学习能力

+5

2025年10月24日

datasetsopen-sourcevision

用 AI Sheets 解锁图像的力量

+2

2025年10月21日

transformerspytorchoptimization

来自OpenAI gpt-oss的技巧，你🫵在transformers中也可以使用

+3

2025年9月11日

spaceszerogpupytorch

ZeroGPU Spaces 加速实践：PyTorch 提前编译全解析

2025年9月2日

openaigptgpt-oss

欢迎 GPT OSS —— 来自 OpenAI 的全新开放模型家族！

+8

2025年8月5日

llmnlpreasoning

SmolLM3: 支持多语言与长上下文推理的小模型

+19

2025年7月8日

smolvlalerobotrobotics

SmolVLA：让机器人更懂'看听说做'的轻量化解决方案

+5

2025年6月3日

nanoVLM: 最简洁、最轻量的纯 PyTorch 视觉-语言模型训练代码库

+3

2025年5月21日

vlmvisionmultimodal

视觉语言模型 (更好、更快、更强)

+1

2025年5月12日

vlmmultimodalvideo

SmolVLM2: 让视频理解能力触手可及

+2

2025年2月20日

AI艺术工具通讯 - 第1期

2025年1月31日

multimodalvlmvision

SmolVLM 越变越小 —— 全新 250 M 和 500 M 模型正式发布！

2025年1月23日

人工智能代理已经到来，接下来呢？

2025年1月13日

multimodalgemmaLLM

欢迎 PaliGemma 2 – 来自 Google 的新视觉语言模型

+1

2024年12月5日

Community Articles

Granite 4.0 Nano: Just how small can you go?

and 1 other •

Why Did MiniMax M2 End Up as a Full Attention Model?

On the Shifting Global Compute Landscape

and 1 other •

The World’s First and Best Speed Painting Software

🛡️ Nemotron PII: Synthesized Data for Privacy-Preserving AI

What makes good reasoning data

NVIDIA Isaac GR00T in LeRobot

and 4 others •

Aligning to What? Rethinking Agent Generalization in MiniMax M2

Cosmos Predict 2.5 & Transfer 2.5: Evolving the World Foundation Models for Physical AI

NVIDIA Releases 8 Million Sample Open Dataset and Tooling for OCR, Image Reasoning, Image and Video QA Tasks

and 6 others •

How to Build a Healthcare Robot from Simulation to Deployment with NVIDIA Isaac for Healthcare

Can Your LLM Think Like a Professional? Introducing ProfBench

and 7 others •

3+ Years of ML & Society at Hugging Face 🤗🤝🧑‍🤝‍🧑

and 3 others •

LightOnOCR-1B: The Case for End-to-End and Efficient Domain-Specific Vision-Language Models for OCR

and 2 others •

Code a simple RAG from scratch

Nemotron-Personas-USA: Synthesized Data for Sovereign AI

and 2 others •

The 1 Billion Token Challenge: Finding the Perfect Pre-training Mix

about 23 hours ago

KV Caching Explained: Optimizing Transformer Inference Efficiency

svara-TTS — Open Multilingual TTS for India’s Voices

Let's talk about LLM evaluation

View all