Blog, Articles, and discussions

Introducing smolagents: simple agents that write actions in code.

By December 31, 2024 • 275

Community Articles

Visualize and understand GPU memory in PyTorch

By December 24, 2024 • 132

Controlling Language Model Generation with NVIDIA's LogitsProcessorZoo

By December 23, 2024 • 30

Evaluating Audio Reasoning with Big Bench Audio

By December 20, 2024 guest • 16

Finally, a Replacement for BERT: Introducing ModernBERT

By December 19, 2024 guest • 450

Bamba: Inference-Efficient Hybrid Mamba2 Model

By December 18, 2024 guest • 37

Welcome the Falcon 3 Family of Open Models!

By December 17, 2024 • 110

Benchmarking Language Model Performance on 5th Gen Xeon at GCP

By December 17, 2024 • 3

Introducing the Synthetic Data Generator - Build Datasets with Natural Language

LeMaterial: an open source initiative to accelerate materials discovery and research

By December 10, 2024 guest • 32

Hugging Face models in Amazon Bedrock

By December 9, 2024 • 10

Open Preference Dataset for Text-to-Image Generation by the 🤗 Community

Welcome PaliGemma 2 – New vision language models by Google

By December 5, 2024 • 123

“How good are LLMs at fixing their mistakes? A chatbot arena experiment with Keras and TPUs

Rethinking LLM Evaluation with 3C3H: AraGen Benchmark and Leaderboard

Community Articles

Blog, Articles, and discussions

Introducing smolagents: simple agents that write actions in code.

Synthetic Data Generation with FastData and Hugging Face

Crowd-sourced Open Preference Dataset for Text-to-Image Generation

Accelerating Language Model Inference with Mixture of Attentions

🌁#82: AI and ML in Real Life

Announcing NVIDIA Cosmos World Foundation Models

How to Automate Reddit Comment Generation with AI Agents in KaibanJS

**Fine-tune SmolLM's on custom synthetic data**

Building Effective Agents with Anthropic’s Best Practices and smolagents ❤️

AI in 2025: A Combinatorial Explosion of Possibilities, but NOT AGI

Superposition in Transformers: A Novel Way of Building Mixture of Experts

**Building a System That Can Build Systems: Toward a Self-Replicating Ecosystem Framework**

Fine-tune a SmolLM on domain-specific synthetic data from a LLM

✴️ ScreenSpot-Pro: GUI Grounding for Professional High-Resolution Computer Use

Process Reinforcement through Implicit Rewards

🐺🐦‍⬛ LLM Comparison/Test: DeepSeek-V3, QVQ-72B-Preview, Falcon3 10B, Llama 3.3 70B, Nemotron 70B in my updated MMLU-Pro CS benchmark

Finetuning Falcon 7b in a hybrid distributed fashion

Debate Championship for LLMs

Fine-tune ModernBERT for text classification using synthetic data

🦸🏻#2: Your Go-To Vocabulary to Navigate the World of AI Agents and Agentic Workflows

Unlocking the Power of Reasoning: Introducing CriticalThinker-LLaMA-3.1-8B-GGUF and Its Groundbreaking Dataset