VibeVoice Technical Report
Paper • 2508.19205 • Published • 180
AI Trends 2026-09
Auto-curated trending papers and models from Hugging Face Hub. Generated by hf-trending-intelligence pipeline. Period: 2026-09-01 to 2026-09-30
Note HOT 0.721 | 2026-09-13 리스코어 (전일 0.766) | upvotes 179 | ���식 모델 87종 (VibeVoice-ASR 720K · Realtime-0.5B 456K DL) | 오픈 토론 20개 | GitHub microsoft/VibeVoice 54.2K★
MinerU2.5: A Decoupled Vision-Language Model for Efficient High-Resolution Document Parsing
Paper • 2509.22186 • Published • 179Note WARM 0.555 | 2026-09-06 | upvotes 177 | 상위 3종 404K DL | 오픈 토론 8개 | 1.2B 문서 파싱 VLM 서빙 표준
SenseNova-U1: Unifying Multimodal Understanding and Generation with NEO-unify Architecture
Paper • 2605.12500 • Published • 198Note WARM 0.662 | 2026-09-06 | upvotes 199 | U1.5-8B-MoT 신규 릴리스(7.8K DL, 212 likes) | 오픈 토론 10개 | GitHub 6.1K★
BDH-CQ: In-Context Learning with Recurrent Latent Reasoning
Paper • 2608.09888 • Published • 795Note WARM 0.406 | 2026-09-12 | upvotes 779 (오늘 1위) | HF 파생 생태계 0 — 구현 lucidrains/bdh-cq 30★, GitHub 검색 22 repo | 150M, ARC-AGI-1 29.5%, $0.0007/task
zai-org/GLM-5.3
Text Generation • 753B • Updated • 1.37M • • 2.02kNote LLM topic-radar trending 1위 | 94403 downloads | Added 2026-09-03 | GLM-5.3 Flash GGUF/Uncensored 파생 동반
SmolDocling: An ultra-compact vision-language model for end-to-end multi-modal document conversion
Paper • 2503.11576 • Published • 177Note Score: 0.484 — WARM | 2026-09-04 | ↑174 | 10 models
Repo-To-Skill: Distilling GitHub Repositories Into AI4AI Skills
Paper • 2609.02749 • Published • 406Note COOL(추적) | 2026-09-06 | upvotes 519 (2위) | Repo-To-Skill: GitHub 레포를 AI4AI 스킬로 증류 | HF 파생 생태계 0, GitHub 정확검색 7레포(커뮤니티 포팅 shuyhere 12★)
SolarWM: Open Data and Scalable Training for Long-Horizon Video World Models
Paper • 2609.02886 • Published • 117Note COOL(추적) | 2026-09-06 | upvotes 139 | SolarWM: 롱호라이즌 비디오 월드모델 오픈 데이터 | 공식 어댑터 6종(Wan2.2-5B/14B·LTX-22B·H3-33B) + SolarWM-Data 4.5K DL
AuK Technical Report: An Open-Source Foundational Model for Speech Generation and Editing
Paper • 2609.08936 • Published • 164Note WARM 0.428 | 2026-09-12 | upvotes 208 | HF models 10 (tencent/AuK) · datasets 8 | GitHub 1 repo | 3.03B instruction-audio 페어, 195만 시간 감독
NeoHorse-1: Towards Recursive Self-Improvement via Agentic Post-Training with Routing Harness
Paper • 2609.08183 • Published • 326Note WARM 0.419 | 2026-09-12 | upvotes 393 | GGUF 파생 확산(mradermacher 8.0K DL) · datasets 0 | GitHub TokenRhythm/NeoHorse 149★
Apodex 1.1: Scaling Agentic Intelligence for Complex Work
Paper • 2608.23283 • Published • 211Note WARM 0.417 | 2026-09-12 | upvotes 206 | GGUF 108K DL · apodex/FrontierChallenge | GitHub 36 repo, AgentHarness 435★
XHToken/Spark-X2.5-4B
Text Generation • 4B • Updated • 39.8k • 1.36kNote LLM topic-radar trending 1위 | 17.7K downloads, 1.1K likes | Added 2026-09-12
deepseek-ai/DeepSeek-V4.1-Flash
Image-Text-to-Text • 763B • Updated • 721k • • 3.94kNote LLM topic-radar trending 1위 | 140.6K downloads, 1,995 likes | Added 2026-09-13
Qwen/Qwen3.8-27B
Image-Text-to-Text • 28B • Updated • 7.04M • • 16.7kNote 전체 trending 상위 | 7.73M downloads, 14.9K likes | GGUF 양자화 파생(unsloth·ISTA-DASLab) 동반 트렌딩 | Added 2026-09-13