Step 3.5 Flash: Open Frontier-Level Intelligence with 11B Active Parameters Paper • 2602.10604 • Published Feb 11 • 193
NVIDIA Nemotron v3 Collection Open, Production-ready Enterprise Models • 15 items • Updated 3 days ago • 223
view article Article Keep the Tokens Flowing: Lessons from 16 Open-Source RL Libraries +7 10 days ago • 66
📋 Twinkle Eval Logs Collection Benchmark log generated with Twinkle Eval, recording the model's outputs for each prompt, see more in https://github.com/ai-twinkle/Eval • 21 items • Updated 6 days ago • 1
🤏 Smol-Data Collection Tried and tested mixes for strong pretraining. Inspired by https://huggingface.co/blog/codelion/optimal-dataset-mixing • 14 items • Updated 18 days ago • 12
view article Article GGML and llama.cpp join HF to ensure the long-term progress of Local AI +4 28 days ago • 488
MMMU-Pro: A More Robust Multi-discipline Multimodal Understanding Benchmark Paper • 2409.02813 • Published Sep 4, 2024 • 33
MMT-Bench: A Comprehensive Multimodal Benchmark for Evaluating Large Vision-Language Models Towards Multitask AGI Paper • 2404.16006 • Published Apr 24, 2024 • 2
view article Article FineWeb-C: A Community-Driven Dataset for Educational Quality Annotations in 122 Languages Jul 8, 2025 • 35
🤗 Fine-zhtw Collection Fine-zhtw is a Traditional Chinese (zh-TW) collection inspired by Hugging Face’s Fine series, built with mostly self-designed methods. • 6 items • Updated Jan 19 • 2