Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models Paper • 2601.07372 • Published Jan 12 • 43
DualPath: Breaking the Storage Bandwidth Bottleneck in Agentic LLM Inference Paper • 2602.21548 • Published 4 days ago • 24
Tri Series Collection Introducing our new series of models: Tri-7B, Tri-21B, and Tri-70B-preview-SFT • 12 items • Updated 9 days ago • 11
AI Release Week Thread (16 February 2026) Collection AI Release Week Thread (16 February 2026) • 10 items • Updated 6 days ago • 1
ColBERT-Zero 🐶 Collection First large-scale fully pre-trained ColBERT model using only public data, outperforming GTE-ModernColBERT and GTE-ModernBERT • 10 items • Updated 9 days ago • 17
VIRTUE Collection Visual-Interactive Text-Image Universal Embedder (ICLR-26) • 5 items • Updated 10 days ago • 3
jina-embeddings-v5-text Collection Our 5th-gen embeddings: two lightweight multilingual models with SOTA performance in retrieval, matching, clustering, and classification. • 29 items • Updated 1 day ago • 31
Ling-2.5 Collection The newest flagship non-reasoning model series. • 1 item • Updated 13 days ago • 8
FireRedASR2S Collection FireRedASR2S is a SOTA, industrial-grade, all-in-one ASR system with ASR, VAD, LID, and Punc module. All modules achieve SOTA performance. • 5 items • Updated 3 days ago • 4
AudioX Collection AudioX is a unified framework for multimodal-conditioned audio and music generation with superior instruction-following capabilities. • 4 items • Updated 16 days ago • 3
pplx-embed Collection Diffusion-Pretrained Dense and Contextual Embeddings • 7 items • Updated 2 days ago • 72