AI models and papers trending on October 14, 2026
The edition of this day as it was recorded: models with the likes and downloads they had then, and the most upvoted papers of the daily list.
On Wednesday, October 14, 2026 the model at the top of the ranking was google/embeddinggemma-2 (feature extraction), with 124,930 downloads in 30 days and 2,133 likes. The most upvoted paper was "Urbex: Agentic Spatial Grounding for City-Scale 3D Scenes" from alibaba, with 50 votes. 12 of the 15 leading papers had a code repository attached.
Recorded 2026-10-14 07:01 UTC · numbers are those of that moment
Papers
- 50
Urbex: Agentic Spatial Grounding for City-Scale 3D Scenes
alibabaarXiv 2610.18356 (opens in a new tab)Code (opens in a new tab)
- 49
Vela: User Behavior as Supervision Sparks LLM Intelligence
AllSpark ResearcharXiv 2610.14709 (opens in a new tab)Code (opens in a new tab)
- 48
DoSA: Dots Sparse Attention
dots studioarXiv 2610.13990 (opens in a new tab)Code (opens in a new tab)
- 38
LeWAM: End-to-End World and Action Modeling with JEPAs
LeWAMarXiv 2610.12818 (opens in a new tab)Code (opens in a new tab)
- 34
Harness^2: Recursive Agent Harnessing for an Open World
GooglearXiv 2610.14224 (opens in a new tab)Code (opens in a new tab)
- 33
Scoring Higher, Answering Worse: Mitigating Reward Hacking in Rubric-Based RL via Protocol-Level Rubrics
ByteDancearXiv 2609.38847 (opens in a new tab)Code (opens in a new tab)
- 31
UltraBERT: Beyond the Limit of BERT Pre-Training
- 30
Multics: Generative Multi-Player World Model with Hierarchical Communication
ARC Lab, TencentarXiv 2610.17201 (opens in a new tab)Code (opens in a new tab)
- 25
Learning How to Remember: Memory Modality Selection for GUI Agents
Korea UniversityarXiv 2610.17928 (opens in a new tab)Code (opens in a new tab)
- 23
TPPO: Tracking-Value PPO For Language Model Reasoning
arXiv 2610.16794 (opens in a new tab)Code (opens in a new tab)
- 22
VideoJev: System One Model for Video Understanding and Interaction
Multimedia Computing Group-Nanjing UniversityarXiv 2610.18615 (opens in a new tab)Code (opens in a new tab)
- 20
SALD: Self-Referenced Advantage Learning for Diffusion Models
Indian Institute of Technology, KanpurarXiv 2610.01496 (opens in a new tab)
- 15
ExeWorldBench: Benchmarking Code-Based Executable 3D World Modeling
University of Science and Technology of ChinaarXiv 2610.15166 (opens in a new tab)
- 14
DeltaTTG: Training Multi-Turn Agents with a Token-to-Go Progress Meter
Nanyang Technological UniversityarXiv 2610.17666 (opens in a new tab)Code (opens in a new tab)
- 13
Jeva: System One Models from First Principles
National University of SingaporearXiv 2610.18773 (opens in a new tab)Code (opens in a new tab)
Models
- 1
Feature extraction · 125K downloads / 30 d · 2,133 likes
- 2
Text generation · 54K downloads / 30 d · 1,548 likes
- 3
Qwen/Qwen-Image-2.1-Turbo (opens in a new tab)
Text to image · 9,665 downloads / 30 d · 726 likes
- 4
abenzerps/Qwen-Image-2.1-Uncensored-Quantized
Text to image · 2.25M downloads / 30 d · 4,232 likes
- 5
openbmb/MiniCPM-V-4.7-35B-A3B (opens in a new tab)
Image text to text · 630 downloads / 30 d · 493 likes
- 6
Image to video · 1.72M downloads / 30 d · 7,671 likes
- 7
ConwayResearch/Underdog-Saluki-27B-1.0 (opens in a new tab)
Text generation · 27K downloads / 30 d · 434 likes
- 8
autotrust/JEV-35B (opens in a new tab)
Text classification · 17.4K downloads / 30 d · 636 likes
- 9
LiquidAI/d1-3B (opens in a new tab)
Image text to text · 13.2K downloads / 30 d · 383 likes
- 10
AtomicChat/Qwen-Image-2.1-Turbo-Uncensored-GGUF (opens in a new tab)
Text to image · 209K downloads / 30 d · 422 likes
- 11
Cloudflare/clef (opens in a new tab)
Image text to text · 17.2K downloads / 30 d · 2,148 likes
- 12
Image text to text · 7.12M downloads / 30 d · 17.7K likes
- 13
Image text to text · 83.9K downloads / 30 d · 312 likes
- 14
baidu/Unlimited-OCR (opens in a new tab)
Image text to text · 1.18M downloads / 30 d · 4,620 likes
- 15
convaiinnovations/laya (opens in a new tab)
Text classification · 46.2K downloads / 30 d · 5,579 likes
About the archive
How far back does the archive go?
Model lists exist from the first day the collector ran: Hugging Face does not publish past trending lists, so they cannot be rebuilt. Paper lists for earlier days were rebuilt from the Hugging Face daily papers archive, and those days are marked "papers only".
What is kept for each day?
The leading models with their task, likes and downloads on that day, the models held back, and the leading papers with their upvotes. Abstracts are not duplicated in the archive; each paper links to arXiv.
Do archived numbers change?
No. A day’s record is the last collection of that day and is not rewritten afterwards. Likes and downloads you see on an archived day are the values on that day, not today’s.
How are weeks defined?
ISO 8601 weeks: Monday to Sunday, numbered so that week 1 contains the first Thursday of the year. The address /week/2026-40 is week 40 of 2026.