view article Article Cosmopedia: how to create large-scale synthetic data for pre-training Large Language Models +1 Mar 20, 2024 • 114
mistralai/Voxtral-Mini-4B-Realtime-2602 Automatic Speech Recognition • 4B • Updated 28 days ago • 890k • 800
Qianfan-VL Collection Qianfan-vl model series. The models are mainly domain enhanced vision language model, targeting enterprise level multi modal understanding scenarios. • 5 items • Updated 21 days ago • 28
💧 LFM2 Collection LFM2 is a new generation of hybrid models, designed for on-device deployment. • 28 items • Updated 25 days ago • 153
High-Resolution Visual Reasoning via Multi-Turn Grounding-Based Reinforcement Learning Paper • 2507.05920 • Published Jul 8, 2025 • 12