view article Article LightOnOCR-1B: The Case for End-to-End and Efficient Domain-Specific Vision-Language Models for OCR By lightonai and 2 others • 4 days ago • 45
Qwen3VL Collection Nexa AI infra to support Qwen3VL running on GPU/NPU/CPU • 22 items • Updated 3 days ago • 3
CoDA Collection CoDA is Salesforce AI Research's open, lightweight and diffusion-based language model. • 2 items • Updated 23 days ago • 5
SINQ: Sinkhorn-Normalized Quantization for Calibration-Free Low-Precision LLM Weights Paper • 2509.22944 • Published about 1 month ago • 76
Hala Technical Report: Building Arabic-Centric Instruction & Translation Models at Scale Paper • 2509.14008 • Published Sep 17 • 87