MemMamba: Rethinking Memory Patterns in State Space Model Paper • 2510.03279 • Published Sep 28 • 72
aquif-Dream Collection Completely open multimodal video generation with audio, built on Wan2.2, Bark Small and SmolVLM. • 1 item • Updated 3 days ago • 1
Less is More: Recursive Reasoning with Tiny Networks Paper • 2510.04871 • Published Oct 6 • 491
aquif-4 Collection aquif-4-Exp is the first hybrid attention model from aquif, built on a strong architecture with 256 experts. • 2 items • Updated 3 days ago • 3
Towards Greater Leverage: Scaling Laws for Efficient Mixture-of-Experts Language Models Paper • 2507.17702 • Published Jul 23 • 6
open1-0 family Collection The first lineup of budget-concious, state-of-the-art open models from Aqui, now with an MIT license. • 6 items • Updated Sep 8 • 1
aquif GGUFs Collection quantized GGUFs for aquif models • 23 items • Updated about 9 hours ago • 1
aquif Models Collection the latest language models from aquif AI: 3.5, 3.6, Grounding, Spatial, DeepResearch • 16 items • Updated about 18 hours ago • 2