Kimi K3 Model Overview: 2.8T Parameters, MXFP4 Quantization, and What the Open Weights Mean for the Community
ResterChed
• • 167
NEW Articles from Team or Enterprise organizations will get promoted to the main section.
Kimi K3 Model Overview: 2.8T Parameters, MXFP4 Quantization, and What the Open Weights Mean for the Community
ResterChed
• • 167The State of Simulation for Physical AI: An Overview
nvidia
• • 66Introducing Cosmos 3 Edge
nvidia
• • 37Hugging Face on AMD Instinct MI455X: First Transformers Results
badaoui
• • 14Be Ready Before the Attack: A Practical Guide to Self-Hosting an Open Model for Cyber Defense
jeffboudier
• • 16POCKET: a 35-billion-parameter model that runs on your iPhone — and on your PC with no GPU
FINAL-Bench
• • 12KV Caching Explained: Optimizing Transformer Inference Efficiency
not-lain
• • 383NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB, Advancing Agentic Retrieval
nvidia
• • 57The influx of specialist models on the Open SLM Leaderboard
Banaxi-Tech
• • 7Uncensor any LLM with abliteration
mlabonne
• • 882Fine-tune video and image models at scale with NVIDIA NeMo Automodel and 🤗 Diffusers
nvidia
• • 80Kimi K3, previewed: inside the first open 3T-class model
embedl
• • 5Introduction to State Space Models (SSM)
lbourdois
• • 238Code a simple RAG from scratch
ngxson
• • 367From GRPO to DAPO and GSPO: What, Why, and How
NormalUhr
• • 133J-Space: Yet Another LLM Mind Reader?
dlouapre
• • 35Tokenization is Killing our Multilingual LLM Dream
omarkamali
• • 17Efficient LLM Pretraining: Packed Sequences and Masked Attention
sirluk
• • 73What is test-time compute and how to scale it?
Kseniase
• • 125Sensitivity Aware Mixed Precision Quantization V1
badaoui
• • 29