Hugging Face Blog
624 posts indexed since 2026
Post Details
| Title | Author | Published | Words | HN Pts |
|---|---|---|---|---|
| Red Teaming with RL: Exploiting Tinker API for Harmful RL on 235B … | weitaofeng | 2026-01-01 | 1,778 | -- |
| The Engineering Handbook for GRPO + LoRA with Verl: Training Qwen2.5 on … | Yağız Çalık | 2026-01-02 | 5,072 | -- |
| Understanding Low-Rank Adaptation (LoRA): A Revolution in Fine-Tuning Large Language Models | Ashish Chadha | 2026-01-03 | 2,023 | -- |
| Introducing Falcon-H1-Arabic: Pushing the Boundaries of Arabic Language AI with Hybrid Architecture | Basma Boussaha, Mohammed Alyafeai, Ahmed Alzubaidi, Leen AlQadi, Shaikha Alsuwaidi, Omar saif alkaabi, Hamza Alobeidli, and Hakim Hacid | 2026-01-05 | 1,838 | 3 |
| TFLOPS Gap: Why FP4 MoE Kernel Engineering Matters on Blackwell | Konstantin | 2026-01-05 | 3,309 | -- |
| Introducing Falcon H1R 7B | Iheb Chaabane, Puneesh Khanna, Suhail M Shah, Slim Frikha, Shi Hu, Abdalgader Abubaker, Reda alami, Mike Lubinets, Mohamed El Amine Seddik, and Hakim Hacid | 2026-01-05 | 1,332 | -- |
| Building Autonomous Vehicles That Reason with the NVIDIA Alpamayo Open Ecosystem | Marco Pavone | 2026-01-05 | 893 | -- |
| NVIDIA Cosmos Reason 2 Brings Advanced Reasoning To Physical AI | Tsung-Yi Lin and Debraj Sinha | 2026-01-05 | 1,037 | -- |
| Scaling Real-Time Voice Agents with Cache-Aware Streaming ASR | Kunal Dhawan, Adi- margolin, Gordana Neskovic, Maryam Motamedi, and Yasmina Benkhoui | 2026-01-05 | 1,860 | -- |
| NVIDIA brings agents to life with DGX Spark and Reachy Mini | Jeff Boudier, Nader Khalil, and Alec Fong | 2026-01-05 | 2,128 | 3 |
| M2.1: Multilingual and Multi-Task Coding with Strong Generalization | MiniMax | 2026-01-05 | 2,306 | -- |
| Generalist Robot Policy Evaluation in Simulation with NVIDIA Isaac Lab-Arena and LeRobot | Raffaello Bonghi, lior ben horin, Kartik S, and Kalyan Vadrevu | 2026-01-05 | 1,038 | -- |
| Small Yet Mighty: Improve Accuracy In Multimodal Search and Visual Document Retrieval … | Ronay Ak, Gabriel de Souza Pereira Moreira, and Bo Liu | 2026-01-06 | 1,492 | -- |
| OpenMed: Six Months of Open-Source Medical AI and the Road Ahead | Maziyar Panahi | 2026-01-06 | 2,424 | -- |
| Why We Built VIBE Bench: Rethinking Evaluation for Real Workloads | MiniMax | 2026-01-06 | 736 | -- |
| Diversity Vs Density: A data strategy comparison for fine-tuning VLMs | Akhil Theerthala | 2026-01-06 | 2,301 | -- |
| 🥃 Distilling Tiny Embeddings | David Mezzetti | 2026-01-10 | 1,082 | -- |
| Proof of Time: A Benchmark for Evaluating Scientific Idea Judgments | Bingyang Ye and Shan Chen | 2026-01-13 | 2,717 | -- |
| SmolLM-Smashed: Tiny Giants, Optimized for Speed | David Berenstein | 2026-01-13 | 982 | -- |
| Reviewer Two (but it's an OpenEnv) | Chris von Csefalvay | 2026-01-13 | 1,653 | -- |
| Introducing OptiMind, a research model designed for optimization | Anson Ho, Sirui Li, and Ishai Menache | 2026-01-15 | 395 | -- |
| How We Built a Semantic Highlight Model To Save Token Cost for … | Cheney Zhang and Jiang Chen | 2026-01-15 | 2,344 | -- |
| Open Responses: What you need to know | shaun smith, ben burtenshaw, merve, and Pedro Cuenca | 2026-01-15 | 1,344 | -- |
| VLM-OCR Recipes on GPU Infrastructure | Florent Gbelidji | 2026-01-15 | 2,281 | -- |
| Beyond Brute Force: Why LoongFlow is the “Thinking” Evolution of OpenEvolve | Xunan Dai | 2026-01-16 | 1,108 | -- |
| LightOnOCR-2-1B: a lightweight high-performance end-to-end OCR model family | Said Taghadouini, Adrien Cavaillès, and Baptiste Aubertin | 2026-01-19 | 934 | -- |
| New in llama.cpp: Anthropic Messages API | Xuan-Son Nguyen and Victor Mustar | 2026-01-19 | 541 | -- |
| A Guide to Reinforcement Learning Post-Training for LLMs: PPO, DPO, GRPO, and … | Karina Zadorozhny | 2026-01-19 | 7,738 | -- |
| Scaling OpenEnv: From Free Usage to Thousands of Concurrent Environments | ben burtenshaw | 2026-01-20 | 1,158 | -- |
| Differential Transformer V2 | Li Dong | 2026-01-20 | 3,136 | -- |
| 🪄 Interpreto: A Unified Toolkit for Interpretability of Transformer Models | Fanny Jourdan and Antonin Poché | 2026-01-20 | 2,112 | -- |
| One Year Since the “DeepSeek Moment” | Adina Yakefu and Irene Solaiman | 2026-01-20 | 1,617 | 4 |
| “DeepSeek R1 时刻” 一周年 | vansin | 2026-01-20 | 315 | -- |
| Waypoint-1: Real-time Interactive Video Diffusion from Overworld | Andrew Lapp, Louis Castricato, Scott Fox, Shahbuland Matiana, and David Rossi | 2026-01-20 | 853 | 92 |
| Security, Governance and Performance for Dell On-Prem AI Builders | Balachandran Rajendran, Juan Julián, Alvaro Bartolome, Enrique Hernández Calabrés, Simon Pagezy, and Jeff Boudier | 2026-01-21 | 1,064 | -- |
| Challenges of Synthetic Dataset Generation | Rishiraj Acharya | 2026-01-21 | 942 | -- |
| AssetOpsBench: Bridging the Gap Between AI Agent Benchmarks and Industrial Reality | Dhaval Patel, James Rayfield, Saumya Ahuja, Chathurangi Shyalika, Shuxin Lin, and Zhou | 2026-01-21 | 1,505 | 3 |
| Optimizing GLM4-MoE for Production: 65% Faster TTFT with SGLang | Novita AI | 2026-01-22 | 1,047 | -- |
| Benchmark Smarter: Tailor Your Model Evaluation Suite with EvalScope | kelseye.xh | 2026-01-22 | 1,973 | -- |
| Reverse Engineering a $500M Mystery: From HashHop to Memory-Augmented Language Models | Asankhaya Sharma | 2026-01-23 | 1,825 | -- |
| RexRerankers: SOTA Rankers for Product Discovery and AI Assistants | Rahul Bajaj, Anuj Garg, and Jaya Nupur | 2026-01-24 | 3,704 | -- |
| Friends and Grandmothers in Silico | Itay Yona | 2026-01-24 | 4,089 | -- |
| Why Your AI Strategy Needs Hugging Face Storage | Adrian Lepers | 2026-01-26 | 1,008 | -- |
| NVIDIA Earth-2 Open Models Span the Whole Weather Stack | Mike Pritchard, Jaideep Pathak, Jean Kossaifi, and Aayush Gupta | 2026-01-26 | 736 | -- |
| The Great Classification Showdown: OSS vs BERT on Consumer Hardware | Ben Toussaint | 2026-01-26 | 1,938 | -- |
| Alyah ⭐️: Toward Robust Evaluation of Emirati Dialect Capabilities in Arabic LLMs | Omar saif alkaabi, Ahmed Alzubaidi, Hamza Alobeidli, Shaikha Alsuwaidi, Mohammed Alyafeai, Leen AlQadi, Basma Boussaha, and Hakim Hacid | 2026-01-27 | 1,585 | -- |
| Unlocking Agentic RL Training for GPT-OSS: A Practical Retrospective | Jason Zhu, Hejian Sang, Arup De, Rohit Jain, and Yanning Chen | 2026-01-27 | 4,160 | -- |
| Architectural Choices in China's Open-Source AI Ecosystem: Building Beyond DeepSeek | Adina Yakefu and Irene Solaiman | 2026-01-27 | 1,324 | 3 |
| Nemotron-Personas-Singapore: Co-Designed Data for Sovereign AI | Will Jennings, Dane Corneil, Yev Meyer, Verdi March, Shyamala Prayaga, and bardiya sadeghi | 2026-01-27 | 1,041 | -- |
| Nemotron-Personas-Brazil: Co-Designed Data for Sovereign AI | Andre Manoel, Yev Meyer, Shyamala Prayaga, Will Jennings, and bardiya sadeghi | 2026-01-28 | 903 | -- |
| We got Claude to teach open models how to write CUDA kernels! | ben burtenshaw, shaun smith, merve, and Pedro Cuenca | 2026-01-28 | 2,350 | 3 |
| Slashing torch.compile Warmup & LoRA Swapping Times with Pruna | John Rachwan, Johanna Sommer, Bertrand Charpentier, and Sara Han Díaz | 2026-01-28 | 1,513 | -- |
| TruthTensor: LLM Evalution in Prediction Markets Under Drift and Market Baseline | Elena Pashkova, shirin Shahabi, Hudson, and Ronald Chan | 2026-01-29 | 1,631 | -- |
| Introducing NVIDIA Cosmos Policy for Advanced Robot Control | Pranjali Joshi, Tsung-Yi Lin, Jinwei Gu, and Prachi Mishra | 2026-01-29 | 1,333 | -- |
| Introducing Daggr: Chain apps programmatically, inspect visually | merve, yuvraj sharma, Abubakar Abid, hysts, and Pedro Cuenca | 2026-01-29 | 1,559 | -- |
| Performant local mixture-of-experts CPU inference with GPU acceleration in llama.cpp | Doctor Shotgun and Geechan | 2026-01-30 | 2,508 | -- |
| Fine-Tuning FunctionGemma on TPU to Create a Virtual Fitness Coach in 10 … | Alvaro Moran | 2026-02-02 | 2,906 | -- |
| Announcing ReasoningLens — Visualizing and Diagnosing LLM Reasoning at a Glance | Jun Zhang, Jason Zheng, Boxi Cao, and ReasoningLens | 2026-02-03 | 693 | -- |
| Training Design for Text-to-Image Models: Lessons from Ablations | David Bertoin, Roman Frigg, and Jon Almazán | 2026-02-03 | 7,420 | -- |
| The Future of the Global Open-Source AI Ecosystem: From DeepSeek to AI+ | Adina Yakefu and Irene Solaiman | 2026-02-03 | 1,602 | 6 |
| H Company's new Holo2 model takes the lead in UI Localization | Ramzi De Coster, Hamza Benchekroun, and Aurélien Lac | 2026-02-03 | 214 | -- |
| Nvidia Agentic Smart Router on Dell Enterprise Hub : Deepdive on Architecture,Design … | Khushboo Rathi and Balachandran Rajendran | 2026-02-03 | 995 | -- |
| 2.2b: FlashAttention — Online Softmax | Atharv Yeolekar | 2026-02-03 | 4,185 | -- |
| Nemotron ColEmbed V2: Raising the Bar for Multimodal Retrieval with ViDoRe V3’s … | Ronay Ak and Gabriel de Souza Pereira Moreira | 2026-02-04 | 1,048 | -- |
| CRAFT: Continuous Reasoning and Agentic Feedback Tuning | Valentin, Denis Timonin, Alexandr, and Alexey | 2026-02-05 | 813 | 5 |
| Introducing SyGra Studio | Surajit Dasgupta, Bidyapati Pradhan, Amit Kumar Saha, Vipul Mittal, and Sriram Puttagunta | 2026-02-05 | 747 | -- |
| 🚀 SyGra V2.0.0 | Sriram Puttagunta, Surajit Dasgupta, Bidyapati Pradhan, Amit Kumar Saha, and Vipul Mittal | 2026-02-05 | 724 | -- |
| From Golden Gate Bridge to Broken JSON: Why Anthropic's SAE Steering Fails … | Maziyar Panahi | 2026-02-07 | 5,766 | -- |
| Building a Mood-Based Movie Recommendation Engine with Voyage-4-nano, Hugging Face, and MongoDB … | Arkadiusz Borucki | 2026-02-08 | 3,315 | -- |
| Transformers.js v4 Preview: Now Available on NPM! | Joshua and Nico Martin | 2026-02-09 | 1,185 | -- |
| Training Qwen3 VL to label bbox : synthetic data, environment and training … | Ulrick BLE | 2026-02-09 | 2,544 | -- |
| Enabling Large Scale RLHF of GPTOSS with Megatron backend in VeRL | LEI WANG | 2026-02-10 | 5,934 | -- |
| 🚀 DTS: A Candidate for the Best Parallel Reasoning in LLMs | Guanchu | 2026-02-11 | 616 | -- |
| OpenEnv in Practice: Evaluating Tool-Using Agents in Real-World Environments | Christian Washington, Ankit Jasuja, Santosh Sah, Lewis Tunstall, and ben burtenshaw | 2026-02-12 | 1,656 | -- |
| LateOn-Code & ColGrep: LightOn unveils state-of-the-art code retrieval models and code search … | Antoine Chaffin and Raphael | 2026-02-12 | 4,993 | 3 |
| How to Use Multiple GPUs in Hugging Face Transformers: Device Map vs … | Aritra Roy Gosthipaty | 2026-02-12 | 606 | -- |
| Compute and Competition in AI: Different FlOPs for Different Folks | Yacine Jernite and Sasha Luccioni | 2026-02-12 | 1,917 | -- |
| Forge: Scalable Agent RL Framework and Algorithm | MiniMax, Hyn, zhi zhang, Jiayuan Song, Da Chen, xkc, Yaoyao, kennyKK, and zpysky1125 | 2026-02-13 | 3,387 | -- |
| Custom Kernels for All from Codex and Claude | ben burtenshaw, Sayak Paul, Aritra Roy Gosthipaty, and shaun smith | 2026-02-13 | 1,792 | 4 |
| What superpower does Kimi-K2.5 bring to the table? | Leco Li | 2026-02-13 | 1,154 | -- |
| The Chinese GLM-5 Model Now Ranks #2 in Arabic Language Performance | Karim Ouda | 2026-02-16 | 322 | -- |
| How to Build a Benchmark with a Private Test Set on Hugging … | Georgia Channing | 2026-02-16 | 1,775 | -- |
| Qwen3.5: Nobody Agrees on Attention Anymore | Maxime Labonne | 2026-02-17 | 1,192 | -- |
| NVIDIA Nemotron 2 Nano 9B Japanese: 日本のソブリンAIを支える最先端小規模言語モデル | Atsunori Fujita, Kotaro Yamamoto, Masaya Ogushi, Vincent Gong, Ameya Sunil Mahabaleshwarkar, and Yoshi Suhara | 2026-02-17 | 297 | -- |
| De-mystifying Multimodal Learning: Enabiling Vision in Language Models | Matteo Nulli | 2026-02-17 | 2,797 | -- |
| DenseR: Dense Rewards For Free in LLM Reasoning | Hritik Bansal | 2026-02-18 | 3,977 | -- |
| One-Shot Any Web App with Gradio's gr.HTML | yuvraj sharma, hysts, and Freddy Boulton | 2026-02-18 | 829 | -- |
| IBM and UC Berkeley Diagnose Why Enterprise Agents Fail Using IT-Bench and … | Ayhan Sebin, Rohan Arora, and Saurabh Jha | 2026-02-18 | 2,253 | -- |
| Did GPT 5.2 make a breakthrough discovery in theoretical physics? | David Louapre | 2026-02-19 | 4,541 | -- |
| ColBERT-Zero: To Pre-train Or Not To Pre-train ColBERT models? | Antoine Chaffin, Luca Arnaboldi, Amélie Chatelain, and Florent Krzakala | 2026-02-19 | 2,306 | -- |
| 「データ不足」の壁を越える：合成ペルソナが日本のAI開発を加速 | Atsunori Fujita, Masaya Ogushi, Will Jennings, Yev Meyer, Kotaro Yamamoto, Yoshi Suhara, Vincent Gong, and Dane Corneil | 2026-02-19 | 280 | -- |
| I Let a Lobster Run My Jetson: What OpenClaw Taught Me About … | Andres Marafioti | 2026-02-19 | 1,509 | -- |
| Train AI models with Unsloth and Hugging Face Jobs for FREE | ben burtenshaw, Daniel (Unsloth), Michael Han, Maxime Labonne, Daniel van Strien, and shaun smith | 2026-02-20 | 944 | -- |
| GGML and llama.cpp join HF to ensure the long-term progress of Local … | Georgi Gerganov, Xuan-Son Nguyen, Aleksander Grygier, Lysandre, Victor Mustar, and Julien Chaumond | 2026-02-20 | 936 | 3 |
| Introducing Legal RAG Bench | Umar Butler and Abdur-Rahman Butler | 2026-02-20 | 3,235 | -- |
| How We Learned to Talk to Machines | Tyler Williams | 2026-02-20 | 1,156 | -- |
| FINAL Bench: The Real Bottleneck to AGI Is Self-Correction | VIDRAFT_LAB | 2026-02-21 | 1,146 | -- |
| GEM Image: Building an AI That Actually Gets Educational Diagrams Right | AIPrep | 2026-02-21 | 966 | -- |
| Kimi K2.5: Still Worth It After Two Weeks? | Maxime Labonne | 2026-02-23 | 1,448 | -- |
| Follow the White Rabbit: Using Embeddings So You Never Get Lost in … | David Corvoysier | 2026-02-23 | 1,420 | -- |
| Do Bubbles Form When Tens of Thousands of AIs Simulate Capitalism? | VIDRAFT_LAB | 2026-02-24 | 2,770 | 5 |
| MAEB: Evaluating Audio Embeddings at Scale | Adnan El Assadi, Solomatin Roman, Kenneth C. Enevoldsen, and Isaac Chung | 2026-02-24 | 1,349 | -- |
| Deploying Open Source Vision Language Models (VLM) on Jetson | Mitesh Patel, Johnny Nuñez Cano, and Raymond Lo | 2026-02-24 | 1,591 | 4 |
| Getting More from Your Test-Time Compute Budget with Portfolio Beam Search | Dan Elbaz, Oren Salzman, Oren Pereg, Daniel Korat, and Ronen Laperdon | 2026-02-24 | 3,527 | -- |
| Visual Aesthetic Benchmark: Can Frontier Models Judge Beauty? | Yichen Feng, Yuetai Li, Chunjiang Liu, Yue Huang, Zhengqing Yuan, Fengqing Jiang, Zichen Chen, and Zhangchen Xu | 2026-02-25 | 1,792 | -- |
| Mixture of Experts (MoEs) in Transformers | Aritra Roy Gosthipaty, Pedro Cuenca, merve, Ilyas Moutawwakil, Arthur Zucker, Sergio Paniego, and Pablo Montalvo | 2026-02-26 | 2,054 | -- |
| Your MoE Model Does Not Have to Select Fixed Number of Experts | Tong Zhu, Xuyang Hu, Xiaoye Qu, Guanjie Chen, and Yu Cheng | 2026-02-26 | 4,405 | -- |
| Bringing Autonomous Driving RL to OpenEnv and TRL | Sergio Paniego | 2026-02-26 | 1,814 | -- |
| AI Coding Assistants Keep Shipping Vulnerable Code -- Here's What We're Doing … | Scott Thornton | 2026-02-26 | 371 | -- |
| A framework and leaderboard for Retrieval Pipelines evaluation on ViDoRe v3 | Quentin Macé, Gabriel de Souza Pereira Moreira, Antoine EDY, Radek Osmulski, and Bo Liu | 2026-02-27 | 1,886 | -- |
| Create, Evaluate, and Connect AI Skills | SkillNet: A Large-Scale Agentic "Skill … | Yuan Liang, Ningyu Zhang, and Xu Ziwen | 2026-02-28 | 2,039 | -- |
| 构建、评估与连接 AI 技能 | SkillNet：大规模智能体“技能图谱”知识库 | Yuan Liang, Ningyu Zhang, and Xu Ziwen | 2026-02-28 | 370 | -- |
| easytranscriber: Speech Recognition with Accurate Timestamps in the HF Ecosystem | Faton Rekathati | 2026-03-03 | 1,169 | -- |
| The ML Engineer's Guide to Protein AI | Maziyar Panahi | 2026-03-03 | 3,612 | -- |
| PRX Part 3 — Training a Text-to-Image Model in 24h! | David Bertoin, Roman Frigg, and Jon Almazán | 2026-03-03 | 1,732 | 5 |
| Introducing Kanon 2 Enricher — the world’s first hierarchical graphitization model | Umar Butler and Abdur-Rahman Butler | 2026-03-03 | 1,571 | -- |
| LLM Architectures Explained: What Powers Today’s Top Models | Sara Han Díaz and Bertrand Charpentier | 2026-03-04 | 1,628 | -- |
| De-mystifying Multimodal Learning: The Hidden Inefficiency in Vision Language Modelling | Matteo Nulli | 2026-03-04 | 2,120 | -- |
| TiRex on the Edge | Robert Weber, Christian Ganhör, and Lukas Fischer | 2026-03-05 | 506 | -- |
| Bringing Robotics AI to Embedded Platforms: Dataset Recording, VLA Fine‑Tuning, and On‑Device … | Gaetan Bahl | 2026-03-05 | 1,851 | -- |
| Introducing Modular Diffusers - Composable Building Blocks for Diffusion Pipelines | YiYi Xu, Alvaro Somoza, Dhruv Nair, and Sayak Paul | 2026-03-05 | 1,907 | 3 |
| NEO-unify: Building Native Multimodal Unified Models End to End | Haiwen Diao, Lewei Lu, and Ziwei Liu | 2026-03-05 | 623 | 3 |
| Building Tucano 2: Open-Source Language Models That Actually Think in Portuguese | Nicholas Kluge Corrêa, Aniket Sen, Shiza Fatimah, Sophia Falk, and Lucie Flek | 2026-03-05 | 2,258 | -- |
| Scaling Pedagogical Pre-training: From Optimal Mixing to 10 Billion Tokens | Asankhaya Sharma | 2026-03-06 | 4,656 | -- |
| Konkani LLM: Bringing a Multi-Script Low-Resource Language to the AI Era | Reuben fernandes | 2026-03-07 | 861 | -- |
| Structural Problems in AI Benchmarking and the Case for a Unified Evaluation … | VIDRAFT_LAB | 2026-03-08 | 1,171 | -- |
| ShopRLVE-GYM: Adaptive Verifiable Environments for E-Commerce Conversational Agents | Rahul Bajaj and Jaya Nupur | 2026-03-08 | 4,976 | -- |
| MARL: Runtime Middleware That Reduces LLM Hallucination Without Fine-Tuning | VIDRAFT_LAB | 2026-03-09 | 1,663 | -- |
| LeRobot v0.5.0: Scaling Every Dimension | Steven Palma, Pepijn Kooijmans, Jade Choghari, Caroline Pascal, Khalil Meftah, Martino Russi, Nicolas Rabault, Michel Aractingi, Virgile BATTO, and Thomas Wolf | 2026-03-09 | 1,931 | -- |
| Granite 4.0 1B Speech: Compact, Multilingual, and Built for the Edge | George Saon and Madison Lee | 2026-03-09 | 385 | -- |
| Ulysses Sequence Parallelism: Training with Million-Token Contexts | Kashif Rasul and Stas Bekman | 2026-03-09 | 3,003 | -- |
| Keep the Tokens Flowing: Lessons from 16 Open-Source RL Libraries | Amine Dirhoussi, Quentin Gallouédec, Kashif Rasul, Lewis Tunstall, Edward Beeching, Albert Villanova del Moral, Nouamane Tazi, and Leandro von Werra | 2026-03-10 | 9,358 | 3 |
| Kanon 2 Reranker: the most powerful reranker for legal RAG | Umar Butler and Abdur-Rahman Butler | 2026-03-10 | 471 | -- |
| How NVIDIA Builds Open Data for AI | Will Jennings, Yev Meyer, Leanna Chraghchian, Rebecca Kao, Jane Polak Scowcroft, and Annie Surla | 2026-03-10 | 1,590 | -- |
| 🏟️ Smol AI WorldCup: A 5-Axis Benchmark That Reveals What Small Language … | VIDRAFT_LAB | 2026-03-10 | 2,482 | 3 |
| Introducing Storage Buckets on the Hugging Face Hub | Lucain Pouget, Eliott Coyac, Adrien Carreira, Victor Mustar, Julien Chaumond, Quentin Lhoest, Pierric Cistac, Sylvestre Bcht, Hugo Larcher, Rajat Arya, Di Xiao, and Assaf Vayner | 2026-03-10 | 1,591 | 18 |
| Code Concepts: A Large-Scale Synthetic Dataset Generated from Programming Concept Seeds | Joseph Jennings and Brandon Norick | 2026-03-11 | 710 | -- |
| Pruna 0.3.2: More OSS Algos, More Ways to Optimize | Minette Kaunismäki, Begüm Çığ, Gaspar Rochette, Sara Han Díaz, and Bertrand Charpentier | 2026-03-11 | 922 | -- |
| How NVIDIA AI-Q Reached #1 on DeepResearch Bench I and II | David Austin | 2026-03-12 | 1,749 | -- |
| Arabic TTS Arena: Ranking Voice Models the Way Chess Ranks Grandmasters | Mohamed Rashad | 2026-03-12 | 1,698 | -- |
| Build an Agent That Thinks Like a Data Scientist: How We Hit … | Jiwei Liu, Maximilian Jeblick, and Jack Yu | 2026-03-13 | 2,052 | -- |
| Beyond Semantic Similarity: Introducing NVIDIA NeMo Retriever’s Generalizable Agentic Retrieval Pipeline | Radek Osmulski, Reza Esfandiarpoor, Yauhen Babakhin, Gabriel de Souza Pereira Moreira, and Bo Liu | 2026-03-13 | 1,520 | -- |
| Super Analyzer: Combining Reasoning and Coding Capabilities to Improve Code Performance | Girish Ganesan and Balachandran Rajendran | 2026-03-13 | 1,363 | -- |
| SILMA TTS: A Lightweight Open Bilingual Text to Speech Model | Karim Ouda | 2026-03-15 | 524 | -- |
| Tokenization is Killing our Multilingual LLM Dream | Omar Kamali | 2026-03-15 | 3,383 | -- |
| The First Healthcare Robotics Dataset and Foundational Physical AI Models for Healthcare … | Sean Huver, Nigel Nelson, Lukas Zbinden, and Mostafa Toloui | 2026-03-16 | 865 | -- |
| Expanding the Alpamayo Open Platform for Developing Reasoning AVs Across Models, Data, … | Marco Pavone | 2026-03-16 | 1,259 | -- |
| NanoVDR: A 70M Text-Only Model That Retrieves Visual Documents as Well as … | Zhuchenyang Liu | 2026-03-16 | 1,493 | -- |
| Canada Must Not Turn AI Chatbots Into a New Surveillance Frontier | Noah Weinberger | 2026-03-16 | 1,934 | -- |
| Holotron-12B - High Throughput Computer Use Agent | Pierre-Louis Cedoz, Hamza Benchekroun, Aurélien Lac, delfosse, Tony Wu, Mats L. Richter, Antoine Bonnet, Kai Yuan, Aleix Cambray (H-AI), and Alexandra | 2026-03-17 | 868 | -- |
| LoRA Fine-Tuning BitNet b1.58 LLMs on Heterogeneous Edge GPUs via QVAC Fabric | Subash SN, Akshay Nambiar, Milan Gritta, Zhen Cong Chen, Arsalan Anwari, Gianfranco Cordella, and Amril Nurman | 2026-03-17 | 3,124 | -- |
| Nemotron 3 Nano 4B: A Compact Hybrid Model for Efficient Local AI | Vinay Raman, Ameya Sunil Mahabaleshwarkar, Hayley Ross, Bilal Kartal, Aditya Malte, Zijia Chen, Ali Taghibakhshi, Sharath Turuvekere Sreenivas, Saurav Muralidharan, Khalil Ben Khaled, Nima Tajbakhsh, Pavlo Molchanov, Oluwatobi Olabiyi, and Yoshi Suhara | 2026-03-17 | 1,552 | 7 |
| State of Open Source on Hugging Face: Spring 2026 | Avijit Ghosh, Lucie-Aimée Kaffee, Yacine Jernite, and Irene Solaiman | 2026-03-17 | 2,883 | -- |
| Pocket Models for iOS: Explore On-Device AI with GGUF Models, Data Memory, … | Hamit Hasanhocaoglu, Arda Dogantemur, Metecan Duyal, and StJohn Deakins | 2026-03-18 | 1,270 | -- |
| Introducing SPEED-Bench: A Unified and Diverse Benchmark for Speculative Decoding | Talor Abramovich, Maor Ashkenazi, Izzy Putterman, Benjamin Chislett, Tiyasa Mitra, Bita Rouhani, Ran Zilberstein, and Yonatan Geifman | 2026-03-19 | 2,333 | -- |
| ATE-2: State-of-the-Art Armenian Text Embeddings and the ArmBench-TextEmbed Benchmark | Hrant Davtyan, Zaruhi Navasardyan, Spartak Bughdaryan, and bag_min | 2026-03-19 | 438 | -- |
| What's New in Mellea 0.4.0 + Granite Libraries Release | Abraham Daniels | 2026-03-20 | 469 | -- |
| Build a Domain-Specific Embedding Model in Under a Day | Steve H, Rucha Apte, Sean Sodha, and Oliver Holworthy | 2026-03-20 | 2,729 | 3 |
| Raw Robot Video to VLA-Ready Training Data: Annotating LeRobot Datasets with Nomadic … | Yunus Cukran | 2026-03-21 | 986 | -- |
| Introducing AI chunking to semchunk | Umar Butler and Abdur-Rahman Butler | 2026-03-23 | 2,228 | -- |
| SynthVision: Building a 110K Synthetic Medical VQA Dataset with Cross-Model Validation | Maziyar Panahi, merve, Jamie@Doubleword, Josh, Seb Ringrose, and Fergus Finn | 2026-03-23 | 3,730 | 3 |
| White Hat Security Agent Prompts 600K Dataset by Yatin Taneja | Yatin Taneja | 2026-03-23 | 1,181 | -- |
| Letter of Superintelligence ~ Yatin Taneja | Yatin Taneja | 2026-03-23 | 1,031 | -- |
| A New Framework for Evaluating Voice Agents (EVA) | Tara Bogavelli, Gabrielle Gauthier Melancon, Katrina Stankiewicz, Nifemi Bamgbose, Hoang Nguyen, Raghav Mehndiratta, Hari Subramani, and Fanny Riols | 2026-03-24 | 2,147 | -- |
| ORBA: Orthogonal Reflection Bounded Ablation — A Geometrically Exact Detour in Directional … | Jim Lai | 2026-03-25 | 5,092 | -- |
| Introducing Cohere-transcribe: state-of-the-art speech recognition | Julian Mack, Ekagra Ranjan, Walter Beller-Morales, Bharat venkitesh, and Pierre Richemond | 2026-03-26 | 1,485 | -- |
| Using Storage Buckets as a Working Layer for Data Pipelines | Daniel van Strien | 2026-03-26 | 1,095 | -- |
| Liberate your OpenClaw 🦀 | Clem 🤗, ben burtenshaw, Pedro Cuenca, Jeff Boudier, merve, Niels Rogge, Victor Mustar, and Mishig Davaadorj | 2026-03-27 | 593 | 4 |
| Introducing WM Bench: A Benchmark for Cognitive Intelligence in World Models | VIDRAFT_LAB | 2026-03-29 | 1,563 | -- |
| How I contributed a new model to the Transformers library using Codex | Niels Rogge | 2026-03-30 | 2,696 | -- |
| 🌈 SKT AI LABS 🌈 | ѕкт αι ℓαвѕ | 2026-03-30 | 555 | -- |
| Training mRNA Language Models Across 25 Species for $165 | Maziyar Panahi | 2026-03-31 | 6,915 | 3 |
| TRL v1.0: Post-Training Library Built to Move with the Field | Quentin Gallouédec, Steven Liu, Pedro Cuenca, and Sergio Paniego | 2026-03-31 | 3,093 | 3 |
| Granite 4.0 3B Vision: Compact Multimodal Intelligence for Enterprise Documents | Madison Lee, Rogerio Feris, Eli Schwartz, Dhiraj Joshi, Pengyuan Li, and Isaac Sanchez | 2026-03-31 | 1,316 | 3 |
| "The Child That Surpassed Both Parents Through MRI-Guided Evolutionary Merge" | VIDRAFT_LAB | 2026-03-31 | 2,884 | -- |
| Falcon Perception | wamiq para and FalconPerception | 2026-04-01 | 2,955 | -- |
| Holo3: Breaking the Computer Use Frontier | Ramzi De Coster, Pierre-Louis Cedoz, Tony Wu, Hamza Benchekroun, mandreux-hai, delfosse, Aurélien Lac, maxime, Axel Moyal, Antonio Loison, Kai Yuan, and Ronan Riochet | 2026-04-01 | 813 | -- |
| Run Gemma 4 on Intel® Arc™ GPUs Out-Of-the-Box | Matrix Yao, Chendi Xue, FanZhao, Xinyu Chen, Alex Gu, Wuxun Zhang, Xinyi Li, jianan, Yi Wang, and Yintong Lu | 2026-04-01 | 1,495 | -- |
| Run Gemma 4 on Intel® Xeon® Out-Of-the-Box | Jiang Li, Xinyu Chen, Chendi Xue, FanZhao, Yi Wang, Wuxun Zhang, Alex Gu, Xinyi Li, jianan, Yintong Lu, and Matrix Yao | 2026-04-01 | 1,464 | -- |
| gradio.Server: Any Custom Frontend with Gradio's Backend | yuvraj sharma and Abubakar Abid | 2026-04-01 | 1,160 | -- |
| Welcome Gemma 4: Frontier multimodal intelligence on device | merve, Pedro Cuenca, Sergio Paniego, ben burtenshaw, Steven Zheng, Alvaro Bartolome, and Nathan Habib | 2026-04-02 | 6,003 | 5 |
| ArmBench-LLM 1.0: Benchmarking LLMs on Armenian Language Tasks | Hrant Davtyan, Zaruhi Navasardyan, Spartak Bughdaryan, and bag_min | 2026-04-02 | 1,205 | -- |
| YC-Bench: Can Your AI Agent Run a Startup Without Going Bankrupt? | Adit, Riddle He, Vincent Tu, Anand Kumar, and Nazneen Rajani | 2026-04-02 | 169 | -- |
| Speculative Decoding in Practice: How EAGLE3 Makes LLMs Faster Without Changing Their … | Gustavo A Lujan and kedar kolluri | 2026-04-03 | 2,730 | -- |
| From doctest to runnable Markdown | Tarek Ziadé | 2026-04-04 | 1,460 | -- |
| How we OCR'ed 30,000 papers using Codex, open OCR models and Jobs | Niels Rogge | 2026-04-07 | 1,246 | 7 |
| BidirLM: Turning Generative LLMs into the Best Open-Source Omnimodal Encoders | Nicolas-BZRD and Théo Deschamps-Berger | 2026-04-07 | 1,772 | -- |
| Darwin V6: Diagnostic-Guided Evolutionary Model Merging | VIDRAFT_LAB | 2026-04-08 | 1,003 | -- |
| Safetensors is Joining the PyTorch Foundation | Luc Georges and Lysandre | 2026-04-08 | 807 | 3 |
| ALTK‑Evolve: On‑the‑Job Learning for AI Agents | Vatche Isahagian, Vinod Muthusamy, Jayaram Radhakrishnan, Gaodan Fang, Punleuk Oum, and G Thomas | 2026-04-08 | 1,180 | 4 |
| Building Harvey-style tabular review from scratch, but better | Abdur-Rahman Butler | 2026-04-09 | 4,508 | -- |
| Multimodal Embedding & Reranker Models with Sentence Transformers | Tom Aarsen | 2026-04-09 | 2,886 | -- |
| Waypoint-1.5: Higher-Fidelity Interactive Worlds for Everyday GPUs | Andrew Lapp, Louis Castricato, Scott Fox, Shahbuland Matiana, and David Rossi | 2026-04-09 | 857 | -- |
| Using OCR models with llama.cpp | Xuan-Son Nguyen | 2026-04-10 | 816 | -- |
| "Darwin-27B-Opus: Surpassing the Foundation Model Without Training" | VIDRAFT_LAB | 2026-04-13 | 1,806 | 4 |
| Releasing LiteCoder-Terminal-SFT | LiteCoder | 2026-04-13 | 833 | -- |
| When Speech AI Meets the Long Tail of Languages: Inside the VAANI … | Sujith Pulikodan, Sanka, Nihar Desai, Suryansh Shukla, and Prasanta Kumar Ghosh | 2026-04-14 | 901 | -- |
| Stop benchmarking inference providers | Nathan Habib | 2026-04-14 | 815 | -- |
| Nucleus-Image: Scaling Text-to-Image with Sparse Mixture of Experts | Nucleus AI | 2026-04-14 | 1,546 | -- |
| Darwin-TTS: We Gave a TTS Model 3% of an LLM's Brain — … | VIDRAFT_LAB | 2026-04-15 | 1,224 | -- |
| Meet HoloTab by HCompany. Your AI browser companion. | Marc Thibault, Pierre-Louis Cedoz, Hamza Benchekroun, Kai Yuan, Aurélien Lac, Tony Wu, Antonio Loison, Axel Moyal, and Emrick Sinitambirivoutin | 2026-04-15 | 516 | -- |
| Inside VAKRA: Reasoning, Tool Use, and Failure Modes of Agents | Ankita Naik, danish, Ben, Anupama Murthi, and Praveen | 2026-04-15 | 3,111 | -- |
| The PR you would have opened yourself | Pedro Cuenca and Awni Hannun | 2026-04-16 | 2,504 | -- |
| easyaligner: Forced alignment of text and audio, made easy | Faton Rekathati | 2026-04-16 | 1,591 | -- |
| Training and Finetuning Multimodal Embedding & Reranker Models with Sentence Transformers | Tom Aarsen | 2026-04-16 | 3,791 | -- |
| Ecom-RLVE: Adaptive Verifiable Environments for E-Commerce Conversational Agents | Rahul Bajaj, Jaya Nupur, Anuj Garg, and ben burtenshaw | 2026-04-16 | 2,563 | -- |
| Building a Fast Multilingual OCR Model with Synthetic Data | Ryan Chesler | 2026-04-17 | 2,218 | 3 |
| NVIDIA Isaac GR00T N1.7: Open Reasoning VLA Model for Humanoid Robots | Edith Llontop and Kalyan Vadrevu | 2026-04-17 | 797 | -- |
| Vessel Browser: The Open Source Browser Designed for Autonomous Agents | Tyler Williams | 2026-04-17 | 845 | -- |
| Multilingual Tool Calling in 70+ Languages, On Device | Bronson, Kato Steven Mubiru, Gimei Alex, OJ Onyeagwu, and Adnan El Assadi | 2026-04-20 | 1,636 | -- |
| Introducing the Bright Data CLI for Automated Web Data Pipelines | Bright Data | 2026-04-20 | 1,786 | -- |
| QIMMA قِمّة ⛰: A Quality-First Arabic LLM Leaderboard | Leen AlQadi, Ahmed Alzubaidi, Mohammed Alyafeai, Maitha Alhammadi, Shaikha Alsuwaidi, Omar saif alkaabi, Basma Boussaha, and Hakim Hacid | 2026-04-21 | 1,731 | -- |
| How to Ground a Korean AI Agent in Real Demographics with Synthetic … | Will Jennings, Hyunwoo Kim, Jinho Lee, jihyeonRyu, Kiran Praveen, Yev Meyer, Kirit Thadaka, and Shyamala Prayaga | 2026-04-21 | 1,502 | -- |
| Save the traces! 🐳 | Pedro Cuenca | 2026-04-21 | 461 | -- |
| DenseOn with the LateOn: Open State-of-the-Art Single and Multi-Vector Models | Raphael Sourty, Antoine Chaffin, Paulo Moura, and Amélie Chatelain | 2026-04-21 | 5,774 | -- |
| AI and the Future of Cybersecurity: Why Openness Matters | Margaret Mitchell, Yacine Jernite, and Clem 🤗 | 2026-04-21 | 1,245 | -- |
| RL: A Structured Human Action & Intent Dataset for Physical AI and … | Gowtham and Marc Hebert | 2026-04-21 | 2,351 | -- |
| mlinter: a linter for Transformers modeling files | Tarek Ziadé | 2026-04-22 | 1,827 | -- |
| Gemma 4 VLA Demo on Jetson Orin Nano Super | Asier Arranz | 2026-04-22 | 1,575 | -- |
| ML Intern Takes Our Post-Training Internship Test | Carlos Miguel Patiño, Aksel Joonas Reedi, and Lewis Tunstall | 2026-04-23 | 924 | -- |
| Hy3 preview: A Rebuilt Hunyuan, a 21B-Active MoE, and a New Reasoning … | Leco Li | 2026-04-23 | 1,035 | -- |
| How to Use Transformers.js in a Chrome Extension | Nico Martin | 2026-04-23 | 1,774 | -- |
| DeepSeek-V4: a million-token context that agents can actually use | ben burtenshaw | 2026-04-24 | 1,488 | 13 |
| Building long-horizon SWE environments on Hugging Face: Frontier SWE × OpenEnv | swappy and Sourasish Basu | 2026-04-26 | 1,224 | -- |
| How to build scalable web apps with OpenAI's Privacy Filter | yuvraj sharma, Freddy Boulton, and Abubakar Abid | 2026-04-27 | 1,641 | -- |
| OpenRA-RL: An Open Platform for AI Agents in Real-Time Strategy Games | Xiaochuang Yuan, huixu, Yiyu Tian, momo, Ruiyue Wang, and Kaiser Sun | 2026-04-27 | 3,015 | -- |
| Running AI agents to automate outreach at scale | Niels Rogge | 2026-04-27 | 2,296 | -- |
| Adaptive Ultrasound Imaging with Physics-Informed NV-Raw2Insights-US AI | Walter Simson, Jay Carlson, Tom Lassiter, Kevin Woo, and Sean Huver | 2026-04-28 | 929 | -- |
| Introducing NVIDIA Nemotron 3 Nano Omni: Long-Context Multimodal Intelligence for Documents, Audio … | Tuomas Rintamaki, Amala Sanjay Deshmukh, Nabin Mulepati, Collin McCarthy, Pritam Biswas, Arushi Goel, Leili Tavabi, Alexandre Milesi, Danial Mohseni Taheri, Kateryna Chumachenko, Isabel Hulseman, Zhehuai Chen, Karan, and Tao | 2026-04-28 | 3,186 | -- |
| BiomedBERT Small: Medical models at 22.7M parameters | David Mezzetti | 2026-04-28 | 912 | -- |
| AI evals are becoming the new compute bottleneck | Avijit Ghosh, Yifan Mai, Georgia Channing, and Leshem Choshen | 2026-04-29 | 3,881 | 4 |
| Pallas for people who know JAX but not kernels yet | Aritra Roy Gosthipaty | 2026-04-29 | 1,581 | -- |
| DeepInfra on Hugging Face Inference Providers 🔥 | Aray Sultanbekova, Shang-Pin, Utemuratov, Yessen K, Oguz Vuruskaner, Célina Hanouti, Simon Brandeis, and Lucain Pouget | 2026-04-29 | 878 | -- |
| Granite 4.1 LLMs: How They’re Built | Yousaf Shah | 2026-04-29 | 2,848 | -- |
| The MCP Era Feels Like Déjà Vu | Mohamed Rashad and Hessah Alharbi | 2026-04-29 | 2,023 | -- |
| Training low-bit ternary models with Axolotl | wing lian | 2026-04-30 | 1,151 | -- |
| Talking to a 4-Year-Old: A Multilingual Benchmark for Children's AI Companions | Batuhan Aktas, Yuvraj, and fatih bugra akdogan | 2026-05-03 | 4,557 | -- |
| Build a legal RAG app that won't be held in contempt | Tabs | 2026-05-05 | 3,115 | -- |
| 🧠 I trained my own French LLM from scratch — alone, with … | vloplok | 2026-05-05 | 2,017 | -- |
| Adding Benchmaxxer Repellant to the Open ASR Leaderboard | Eric Bezzam, Steven Zheng, Eustache Le Bihan, Sergio Bruccoleri, Jeanine Sinanan-Singh, Casey Ford, Guanbo Wang, Yukai Huang, Ke Li, Yufeng Hao, and Liao Xiaoling | 2026-05-06 | 1,400 | -- |
| Learning Maths for the Last Time | Shane, LaneFiedler, Enderchef (Enderchefcoder), LH-Tech AI, Arman Rafiee, poe, and AxionLab | 2026-05-06 | 1,325 | -- |
| Introducing the agentic robotics appstore for 10,000 Reachy Minis | Clem 🤗 | 2026-05-06 | 1,207 | -- |
| vLLM V0 to V1: Correctness Before Corrections in RL | Rafael Pardinas and Ehsan Kamalloo | 2026-05-06 | 1,579 | -- |
| QVAC MedPsy: State-of-the-Art Medical and Healthcare Language Models for Edge Devices | Mathias Buus, Davide Vitabile, Alex Buffa, Akshay Nambiar, and Amril Nurman | 2026-05-07 | 9,495 | 3 |
| Improving Depth Anything V2 Robustness to Video Compression | Ethan F and Ronen Nissim | 2026-05-07 | 3,407 | -- |
| MedQA: Fine-Tuning a Clinical AI on AMD ROCm — No CUDA Required | Harikrishna | 2026-05-08 | 1,520 | -- |
| EMO: Pretraining mixture of experts for emergent modularity | Kyle Wiggers and Ryan Wang | 2026-05-08 | 1,830 | -- |
| CyberSecQwen-4B: Why Defensive Cyber Needs Small, Specialized, Locally-Runnable Models | Samuel | 2026-05-08 | 1,783 | -- |
| Hugging Face on JFrog Artifactory: An Enterprise Guide (and What Changes in … | Jeff Boudier | 2026-05-08 | 5,080 | -- |
| "OncoAgent: A Dual-Tier Multi-Agent Framework for Privacy-Preserving Oncology Clinical Decision Support" | Máximo López Chenlo | 2026-05-09 | 2,938 | -- |
| Building Blocks for Foundation Model Training and Inference on AWS | Keita Watanabe, Pavel Belevich, and Aman Shanbhag | 2026-05-11 | 4,362 | -- |
| Two Years of Local AI on a Laptop: When Open Models Outpaced … | Mishig Davaadorj | 2026-05-11 | 1,653 | 3 |
| Safety Evals Should Project Test-Time Compute | Tommaso Cerruti | 2026-05-11 | 2,521 | -- |
| You do the work. Big Tech takes the model. | Urro | 2026-05-11 | 3,960 | -- |
| Self Evolving is the Endgame or final destiny | Rajkumar rawal | 2026-05-12 | 683 | -- |
| Vocabulary-Augmented Prompting for Sango — Production African Language AI Without a Parallel … | MICWEN | 2026-05-13 | 3,112 | -- |
| The Best Open Source and Open-Weight LLM Models to Run Locally in … | Daya Shankar | 2026-05-13 | 4,740 | -- |
| Unlocking asynchronicity in continuous batching | Rémi Ouazan Reboul, Pedro Cuenca, and Aritra Roy Gosthipaty | 2026-05-14 | 4,015 | -- |
| How to Comply with SOC 2 and ISO 27001 with Hugging Face: … | Jeff Boudier | 2026-05-14 | 3,007 | -- |
| Granite Embedding Multilingual R2: Open Apache 2.0 Multilingual Embeddings with 32K Context … | Radu Florian, Parul Awasthy, Aashka Trivedi, and Madison Lee | 2026-05-14 | 3,411 | -- |
| Training-Free Reasoning at 88.89% on GPQA Diamond: How Darwin Family Hit Frontier … | VIDRAFT_LAB | 2026-05-15 | 882 | -- |
| Vividh-ASR: Diagnosing and Fixing Studio-Bias in Whisper for Indic Languages | Kavya Manohar, Kush Juvekar, and Kumarmanas Nethil | 2026-05-15 | 3,877 | -- |
| AI, Physical AI, World Models, VLA, VLM, and Other Terms We Should … | Barak Or | 2026-05-17 | 2,009 | -- |
| PaddleOCR 3.5: Running OCR and Document Parsing Tasks with a Transformers Backend | AlexZhang, cuicheng, Jun Zhang, and Manhui Lin | 2026-05-18 | 927 | -- |
| Fine-Tuning NVIDIA Cosmos Predict 2.5 with LoRA/DoRA for Robot Video Generation | Ting-Yun Chang, Miguel Martin, Jonathan Allen, Ke Ding, and Pooya Jannaty | 2026-05-18 | 2,653 | -- |
| The Open Agent Leaderboard | Elron Bandel | 2026-05-18 | 1,703 | 3 |
| OlmoEarth v1.1: A more efficient family of models | Kyle Wiggers | 2026-05-19 | 898 | -- |
| Introducing the Ettin Reranker Family | Tom Aarsen | 2026-05-19 | 5,698 | -- |
| Software Forgets: Agent Traces Are the Memory | Caleb Fahlgren | 2026-05-19 | 604 | -- |
| SANA-WM Bidirectional on Apple Silicon | Arjun Reddy | 2026-05-20 | 1,105 | -- |
| LeRobot Humanoid: An Open, Low-Cost, 3D-Printed Humanoid for Robot Learning | Virgile BATTO, Caroline Pascal, Steven Palma, Maxime Ellerbach, Nicolas Rabault, Martino Russi, and haixuan tao | 2026-05-21 | 1,550 | -- |
| Specialization Beats Scale: A Strategic Variable Most AI Procurement Decisions Overlook | Erick Lachmann and Pimenta de Freitas Cardoso | 2026-05-22 | 2,753 | -- |
| Why Open Models Are the Only Sustainable Way to Teach AI | Pénélope Gittos | 2026-05-22 | 1,325 | -- |
| Eight Days in China: What I Learned from the AI Labs, Robotics … | Matt White | 2026-05-22 | 12,170 | -- |
| Towards Speed-of-Light Text Generation with Nemotron-Labs Diffusion Language Models | Mehran Maghoumi, Yonggan Fu, Pavlo Molchanov, and Khadkevich | 2026-05-23 | 1,167 | -- |
| An experiment with attention. | poe, Lane Fiedler, Shane, and Enderchef (Enderchefcoder) | 2026-05-23 | 1,061 | -- |
| Relaunching PapersWithCode with new features | Niels Rogge | 2026-05-24 | 498 | -- |
| Harness, Scaffold, and the AI Agent Terms Worth Getting Right | Sergio Paniego and Aritra Roy Gosthipaty | 2026-05-25 | 2,117 | -- |
| Borealis — open data, code, weights recipe for training Audio LLM | Wortega | 2026-05-25 | 2,303 | -- |
| Should we use genetics instead of system prompts for AI Agents & … | Fyx | 2026-05-25 | 2,550 | -- |
| Give your agents ZeroGPU to ship viral AI apps autonomously | Victor Mustar | 2026-05-26 | 941 | -- |
| ITBench-AA: Frontier Models Score Below 50% on the First Benchmark for Agentic … | Ayhan Sebin, Saurabh Jha, and Rohan Arora | 2026-05-27 | 889 | -- |
| Reachy Mini goes fully local | Amir Mahla and Andres Marafioti | 2026-05-27 | 1,849 | 4 |
| Shipping a Trillion Parameters With a Hub Bucket: Delta Weight Sync in … | Amine Dirhoussi, Quentin Gallouédec, Kashif Rasul, Lewis Tunstall, Edward Beeching, Albert Villanova del Moral, and Leandro von Werra | 2026-05-27 | 4,227 | -- |
| Server is at capacity | specimba, Lewis Tunstall, and Aksel Joonas Reedi | 2026-05-27 | 266 | -- |
| Introduction to Trimming ✂ | Loïck BOURDOIS, Tom Aarsen, Bram Vanroy, Woojun Jung, Manuel Romero, and Prithiv Sakthi | 2026-05-28 | 19,577 | -- |
| MONET: Lowering the bar for World-Class Image Generation research. | Benjamin Aubin, Gonzalo Quintana, Onur, sanjeev sreetharan, Czerwinska, Damien Henry, and Clément Chadebec | 2026-05-28 | 1,601 | -- |
| Profiling in PyTorch (Part 1): A Beginner's Guide to torch.profiler | Aritra Roy Gosthipaty, Sayak Paul, Sergio Paniego, Rémi Ouazan Reboul, and Pedro Cuenca | 2026-05-29 | 5,132 | -- |
| Dell Enterprise Hub at Dell Tech World 2026: new models, new platforms, … | Simon Pagezy, Enrique Hernández Calabrés, Juan Julián, Bagus Hanindhito, Girish Ganesan, ravikumar, Ian Roche, Jeff Boudier, and Balachandran Rajendran | 2026-05-29 | 1,112 | -- |
| Agentic RL: Token-In, Token-Out Done Right | Quentin Gallouédec and Kashif Rasul | 2026-05-29 | 3,670 | -- |
| MiniMax Goes Sparse: Decoding M3's Attention from a Single Diagram | Atlas Cloud | 2026-05-29 | 1,680 | -- |
| Does Depth Actually Help Reasoning? A Tiny Experiment on 2× T4 | Wop and Lane Fiedler | 2026-05-30 | 708 | -- |
| ClawHub Security Signals: Large Corpus Multi-Scanner Dataset for Agent Skill Security Research | Vincent Koc, Patrick Erichsen, Jacob Tomlinson, Agustin Rivera, Mike Appel, and Nir Paz | 2026-06-01 | 1,400 | -- |
| Welcome NVIDIA Cosmos 3: The First Open Omni-model for Physical AI Reasoning … | Asawaree and Atharva Joshi | 2026-06-01 | 1,960 | -- |
| Beyond LLMs: Why Scalable Enterprise AI Adoption Depends on Agent Logic | Nicholas Fuller | 2026-06-01 | 2,177 | -- |
| A Deep Neural Network that turns Any Image into a Playable Game! … | Abhishek Sensharma | 2026-06-01 | 365 | -- |
| Introducing Mellum2: A 12B Mixture-of-Experts Model by JetBrains | Nikita Pavlichenko | 2026-06-01 | 600 | -- |
| Taking Alpamayo to New Heights with Driving Foundation Models and Closed-Loop Training | Marco Pavone and Boris Ivanovic | 2026-06-01 | 1,386 | -- |
| From Data Repositories to Production Data Pipelines: Bridging Hugging Face Datasets and … | Parag Ekbote | 2026-06-01 | 1,357 | -- |
| Taking Alpamayo to New Heights with Driving Foundation Models and Closed-Loop Training | Marco Pavone and Boris Ivanovic | 2026-06-01 | 1,380 | -- |
| Holo3.1: Fast & Local Computer Use Agents | Maxime Langevin, Hamza Benchekroun, Axel Moyal, Emrick Sinitambirivoutin, Antonio Loison, Avshalom Manevich, Tony Wu, Pierre-Louis Cedoz, Aurélien Lac, and Ronan Riochet | 2026-06-02 | 867 | -- |
| AutoResearch on Diffusers' Pipeline for 10 Rounds on JarvisLabs | chansung park | 2026-06-03 | 2,294 | -- |
| Adding MCP Tools to Reachy Mini | Alina Lozovskaya | 2026-06-03 | 2,068 | -- |
| Direct Preference Optimization Beyond Chatbots | Erick Lachmann and Pimenta de Freitas Cardoso | 2026-06-03 | 2,953 | 3 |
| How to Fine-Tune Nemotron 3.5 ASR for Your Language, Domain, or Accent | Maryam Motamedi, Adi- margolin, Francesco, Myungjong Kim, Enas Albasiri, and Jinhan Wang | 2026-06-04 | 2,254 | -- |
| Fine-tune FLUX.2 [klein] with a LoRA under 60 minutes | Stephen Batifol | 2026-06-04 | 2,617 | -- |
| EVA-Bench Data 2.0: 3 Domains, 121 Tools, 213 Scenarios | Tara Bogavelli, Gabrielle Gauthier Melancon, Katrina Stankiewicz, Nifemi Bamgbose, Fanny Riols, Hoang Nguyen, Raghav Mehndiratta, Lindsay Brin, Joseph Marinier, Hari Subramani, and Anil Madamala | 2026-06-04 | 1,990 | -- |
| Task-Seeded Synthetic Q&A Generation for Nemotron Pretraining | Dan Su | 2026-06-04 | 1,811 | -- |
| Designing the hf CLI as an agent-optimized way to work with the … | Célina Hanouti and Lucain Pouget | 2026-06-04 | 2,856 | -- |
| Nemotron 3.5 Content Safety: Customizable Multimodal Safety for Global Enterprise AI | Varun Singh , Isabel Hulseman, Anuj Doshi, and Shyamala Prayaga | 2026-06-04 | 2,226 | -- |
| Build Small Hackathon With Cohere Models | Alejandro Rodriguez | 2026-06-04 | 2,018 | -- |
| Thousand Token Wood: shipping a multi-agent economy on a 3B model | Lester Leong | 2026-06-05 | 1,029 | -- |
| Job Searcher | Emre | 2026-06-06 | 872 | -- |
| Five labs, five minds: building a multi-model finance drama on small models | Lester Leong | 2026-06-06 | 1,141 | -- |
| Run Claude Code, OpenCode & Frontier Coding Models on Your Own AI … | Girish Ganesan and Balachandran Rajendran | 2026-06-06 | 1,723 | -- |
| Her · हेर — a detective for your Claude Code sessions | Ashish Chalke | 2026-06-07 | 622 | -- |
| Mobile Manipulation with LeKiwi and PincOpen | Xingdong Zuo | 2026-06-07 | 3,431 | -- |
| Building Pakistan Notice Helper: A Small AI Tool for a Very Local … | Abid Ali Awan | 2026-06-08 | 2,724 | -- |
| The Open Source Community is backing OpenEnv for Agentic RL | ben burtenshaw, Joseph Spisak, Lysandre, Davide Testuggine, will brown, Joy Liu, Peyton Walters, Chris Wing, Daniel (Unsloth), Andrew Zhou, Michael Han, Hamid Shojanazeri, Sanyam Bhutani, Zach Wentz, Emre Guven, Lewis Tunstall, and Sergio Paniego | 2026-06-08 | 850 | -- |
| Arcee Becomes the First Major American AI Lab to Replace AWS S3 … | Clem 🤗, Lucas Atkins, and Mark McQuade | 2026-06-09 | 900 | -- |
| How an Agent Built a 3D Paris Gallery by Chaining Two Hugging … | Mishig Davaadorj | 2026-06-09 | 907 | -- |
| Can Voice Agents Handle Bilingual Customers? Benchmarking Frontier ASR on Code-Switched Speech | Shama Gupta, Lindsay Brin, and Fanny Riols | 2026-06-09 | 2,621 | -- |
| Migrating Your GitHub CI to Hugging Face Jobs | Abubakar Abid | 2026-06-09 | 1,753 | -- |
| Introducing North Mini Code: Cohere’s First Model For Developers | Cohere Code Agents Team | 2026-06-09 | 2,737 | -- |
| NeuroBait: I fine-tuned a model to spark dopamine for ADHD brain | Harisabekti Dicky Subrata | 2026-06-09 | 794 | -- |
| Eyes, ears, and a voice: building Reachy Mini's media stack | Fabien Danieau, Alina Lozovskaya, Caroline Pascal, and Antoine Pirrone | 2026-06-10 | 2,694 | -- |
| 36 Prompts, One Infinite City | Mishig Davaadorj | 2026-06-10 | 805 | -- |
| Lolaby — AI-powered lullabies | André Oliveira and Vasco Oliveira | 2026-06-11 | 1,484 | -- |
| Profiling in PyTorch (Part 2): From nn.Linear to a Fused MLP | Aritra Roy Gosthipaty, Rémi Ouazan Reboul, Sergio Paniego, Pedro Cuenca, and Sayak Paul | 2026-06-11 | 3,681 | -- |
| PhysicsIntern: from an Autonomous Benchmark-runner to a Research Sidekick | David Louapre | 2026-06-11 | 2,125 | -- |
| Optimum Intel 2.0: An OpenVINO-First Toolkit for Running Open Models on Intel | Jeff Boudier and Ella Charlaix | 2026-06-11 | 1,038 | -- |
| MTEB Leaderboard: From a slow demo to feature-rich leaderboard | Solomatin Roman, Kenneth C. Enevoldsen, and Isaac Chung | 2026-06-12 | 955 | -- |
| olmo-eval: An evaluation workbench for the model development loop | Tyler Murray and Kyle Wiggers | 2026-06-12 | 1,545 | -- |
| Introducing Serge: GitHub-Native AI Code Review | Tarek Ziadé and Sayak Paul | 2026-06-12 | 1,443 | -- |
| FINAL-Bench Quantum: An Open, Neutral Benchmark for Quantum-Computing Methods | VIDRAFT_LAB | 2026-06-14 | 679 | -- |
| PitchFight AI: Practice the Pitch Before the Real Room | Prakhar Parashar | 2026-06-14 | 743 | -- |
| Closet Twin: Your AI-Powered Personal Stylist Built for the Build Small Hackathon | Nouhaila mfth | 2026-06-14 | 694 | -- |
| Building an AI Interview Coach for the BuildSmall Hackathon 2026 | Ishan Awasthi | 2026-06-15 | 1,163 | -- |
| Eyas: AI Security Camera Agent | Seunghyun(Joe), Hanhee Lee, and Javier Huang | 2026-06-15 | 1,983 | -- |
| QLORA SFT Distillation Effects on Qwen3.6 27B Agentic Coding Harness Fluency | Thomas Kim | 2026-06-15 | 1,939 | -- |
| The Office Meets Silicon Valley | Felix | 2026-06-15 | 1,709 | -- |
| How We Built OpenMythos: A Cybersecurity LLM Trained from Scratch | Nishith Jain | 2026-06-15 | 1,139 | -- |
| I fine-tuned a model for free from one prompt, with TRL and … | Sergio Paniego | 2026-06-15 | 922 | -- |
| 🧬 Carbon-VEPor: Efficient Variant Effect Prediction with Carbon | Vivek Silimkhan | 2026-06-15 | 1,743 | -- |
| Party is over: regularizing ColBERT models to fix efficient ANN methods | Antoine Chaffin | 2026-06-16 | 5,150 | -- |
| GLM-5.2: Built for Long-Horizon Tasks | Z.AI | 2026-06-17 | 2,853 | 10 |
| From the Hugging Face Hub to robot hardware with Strands Agents and … | Sundar Raghavan and Cagatay Cali | 2026-06-17 | 3,491 | -- |
| MolmoMotion: Language-guided 3D motion forecasting | Kyle Wiggers | 2026-06-17 | 1,901 | -- |
| Intel XPU Kernel Skill: LLM-driven Triton kernel optimization for the Hugging Face … | Daniel Fleischer and Moshe Wasserblat | 2026-06-17 | 2,201 | -- |
| Agentic Resource Discovery: Let agents search for tools, skills, and other agents. | ben burtenshaw and shaun smith | 2026-06-17 | 1,418 | -- |
| MosaicLeaks: Can your research agent keep a secret? | Alexander Gurung and Rafael Pardinas | 2026-06-18 | 1,889 | -- |
| Is it agentic enough? Benchmarking open models on your own tooling | Lysandre, Nathan Habib, and Pedro Cuenca | 2026-06-18 | 3,363 | -- |
| Beyond LoRA: Can you beat the most popular fine-tuning technique? | Benjamin Bossan, Sayak Paul, Marian, and Kashif Rasul | 2026-06-18 | 2,754 | -- |
| Enterprise AI benchmarks: head-to-head comparison of Falconer, Notion, Atlassian Rovo, Claude Code, … | Maximiliano Benedetto and Matt Zhao | 2026-06-18 | 1,668 | -- |
| No Photoshop, No Blender: Multimedia by Agent | Mishig Davaadorj | 2026-06-19 | 1,166 | -- |
| Continuous batching for GRPO, now in TRL | Sergio Paniego | 2026-06-19 | 712 | -- |
| Where Does the Signal Live? A Web Data Recipe for Medical Encoder … | bofeng huang, Sun Jacques, Diane Bouchacourt, Nicolas Barascud, and Fajwel Fogel | 2026-06-20 | 2,019 | -- |
| PP-OCRv6 on Hugging Face: 50-Language OCR from 1.5M to 34.5M Parameters | AlexZhang, cuicheng, Jun Zhang, Manhui Lin, Yue Zhang, leo-q8, yubo, and Yi Liu | 2026-06-22 | 1,089 | -- |
| We got local models to triage the OpenClaw repo for FREE!* | Onur Solmaz, ben burtenshaw, shaun smith, Pedro Cuenca, and Lysandre | 2026-06-22 | 2,891 | -- |
| V-Zero | haoxiang sun | 2026-06-22 | 859 | -- |
| Shipping huggingface_hub every week with AI, open tools, and a human in … | Lucain Pouget and Célina Hanouti | 2026-06-23 | 2,343 | -- |
| Experimenting with the proposed Cross-Origin Storage API in Transformers.js | Thomas Steiner | 2026-06-23 | 2,915 | -- |
| Build real agentic apps using CUGA: two dozen working examples on a … | Anupama Murthi, Hamid Adebayo, Sami Marreed, Praveen, and Asaf Adi | 2026-06-23 | 3,392 | -- |
| Machine learning for alien climates: Introducing the ThousandWorlds benchmark | Edward Stevenson | 2026-06-23 | 899 | -- |
| Building Moon Bot: A Slack-Native Coding Agent Backed by HuggingFace Buckets | Eliott Coyac, Caleb Fahlgren, and Franck Abgrall | 2026-06-24 | 2,003 | -- |
| Interhuman’s Goblin: “Yeah, Friday at Five” | Siddharth Ravi | 2026-06-24 | 2,371 | -- |
| Accelerating Transformers Fine-Tuning with NVIDIA NeMo AutoModel | Adil Asif, Alexandros Koumparoulis, Wenwen Gao, Sylendran Arunagiri, David Messina, and Bernard Nguyen | 2026-06-24 | 2,234 | -- |
| Introducing the FFASR Leaderboard: Benchmarking ASR in the Real World | Daniel Gert Nielsen, Shivam Saini, Alessia Milo, Georg Götz, and Eric Bezzam | 2026-06-24 | 1,647 | -- |
| Kog Laneformer 2B: The Latency-First Model Behind Kog Inference Engine | Morgan Giraud, Gauthier Tallec, and Gaël Delalleau | 2026-06-24 | 3,042 | -- |
| Which tokens does a hybrid model predict better? | Kyle Wiggers | 2026-06-25 | 1,364 | -- |
| Run a vLLM Server on HF Jobs in One Command | Quentin Gallouédec | 2026-06-26 | 1,611 | -- |
| VLX-Flow: Continuous Video Understanding for Real-Time Multimodal Interaction | Tony Zhao and Yibo Ma | 2026-06-26 | 1,194 | -- |
| OlmoLogic: Boosting Reasoning via RLVR with Inductive Logic Programming | Lukas Helff, Sebastian Sztwiertnia, Felix Friedrich, Hikaru Shindo, and Ahmad Omar | 2026-06-26 | 2,702 | -- |
| SportsBERT Small: Domain-specialized small models | David Mezzetti | 2026-06-26 | 1,126 | -- |
| VLX-Flow: Continuous Video Understanding for Real-Time Multimodal Interaction | Tony Zhao | 2026-06-27 | 1,223 | -- |
| VLX-Seek: Improving VLM Fine-Grained Perception via Region Reference Instead of Coordinate Generation | Peng Liu and Tony Zhao | 2026-06-27 | 3,375 | -- |
| DukaanBench: Can AI Run an Indian Grocery Store for 30 Days? | Ekansh Srivastva | 2026-06-27 | 3,871 | -- |
| VLX-Go: Vision-Language Short-Horizon Waypoint Prediction for Embodied Navigation | Peng Liu and Tony Zhao | 2026-06-28 | 1,138 | -- |
| Chitos: From Detection to Proof — An Autonomous Security AI That Actually … | VIDRAFT_LAB | 2026-06-29 | 1,660 | -- |
| 80TB+ of astronomy for the HDD-poor: crossmatch the Multimodal Universe from your … | Mike Smith | 2026-06-29 | 2,494 | -- |
| DiScoFormer: One transformer for density and score, across distributions | Kyle Wiggers | 2026-06-29 | 894 | -- |
| Gemma-4 31B + vLLM on RTX 6000 PRO : A Real-Load Benchmark | Nikhil K. | 2026-06-29 | 786 | -- |
| Featuring Every Eval Ever Results on Hugging Face Model Pages | Sree Harsha Nelaturu, Avijit Ghosh, Nathan Habib, Jan Batzner, Leshem Choshen, Irene Solaiman, and Julien Chaumond | 2026-06-30 | 1,434 | -- |
| Why Specialization Is Inevitable | Erick Lachmann and Francisco de Almeida Rocha Alves | 2026-06-30 | 2,264 | -- |
| ScarfBench: Benchmarking AI Agents for Enterprise Java Framework Migration | Raju Pavuluri, Rahul Krishna, Srikanth Govindaraj Tamilselvam, Bridget M, Ashita Saxena, George Safta, Advait Pavuluri, and Michele Merler | 2026-06-30 | 1,067 | -- |
| Heretic Grimoire | Vinay Umrethe | 2026-06-30 | 1,360 | -- |
| Does Your LLM Know When It's About to Be Wrong? | ginigen-ai | 2026-07-01 | 1,446 | -- |
| Hugging Face and Cerebras bring Gemma 4 to real-time voice AI | Amir Mahla, Andres Marafioti, Leandro von Werra, and Saurabh Vyas | 2026-07-01 | 576 | -- |
| Pulpie: Pareto-Optimal Models for Cleaning the Web | Shreyash Nigam | 2026-07-01 | 2,232 | -- |
| AstroBERT Small: Domain-specialized small models | David Mezzetti | 2026-07-01 | 1,249 | -- |
| Claude Fable 5 — Technical Harness Report | NIONGOLO Chrys Fé-Marty | 2026-07-01 | 3,626 | -- |
| BaseRT: Best-in-Class LLM Inference on Apple Silicon via Native Metal | Prabod, Fabian Waschkowski, and Lukas Wesemann | 2026-07-01 | 1,290 | -- |
| Adding a GPU Without Building One | VIDRAFT_LAB | 2026-07-03 | 1,374 | -- |
| 🔁 Apprendre à un LLM français de 15M à penser plus profond … | RDTvlokip | 2026-07-03 | 5,522 | -- |
| 🔁 Teaching a 15M French LLM to think deeper — and to … | RDTvlokip | 2026-07-03 | 4,769 | -- |
| Quantum Cryptanalysis on Real Hardware: Pushing Symmetric-Structure Key Recovery Beyond the Published … | VIDRAFT_LAB | 2026-07-05 | 2,183 | -- |
| Distilling OmniVoice into Aegis: Female Urdu TTS at 61 MB ONNX for … | Mahwiz Khalil | 2026-07-05 | 1,151 | -- |
| 🤗 Kernels: Major Updates | Sayak Paul, Daniël de Kok, and David Holtz | 2026-07-06 | 1,812 | 5 |
| PRX Part 4: Our Data Strategy | Roman Frigg, David Bertoin, and Jon Almazán | 2026-07-06 | 4,298 | -- |
| After the party comes the free lunch: regularizing ColBERT models to enhance … | Antoine Chaffin | 2026-07-06 | 2,819 | -- |
| LeRobot v0.6.0: Imagine, Evaluate, Improve | Steven Palma, Pepijn Kooijmans, Caroline Pascal, Khalil Meftah, Martino Russi, Nikodem Bartnik, Nicolas Rabault, and Thomas Wolf | 2026-07-07 | 2,614 | -- |
| Run AI workloads on any cloud, store on Hugging Face: zero-egress storage … | Nikhil Jha, Zhanghao Wu, Hope Wang, Adrien Carreira, and Julien Chaumond | 2026-07-07 | 1,818 | -- |
| Teaching a coding agent to deploy production endpoints on Amazon SageMaker | Dario Salvati, Alvaro Bartolome, and Jeff Boudier | 2026-07-07 | 3,582 | -- |
| Hugging Face Models on Foundry Managed Compute | Manoj Bableshwar and Osi | 2026-07-07 | 2,222 | -- |
| From Hugging Face to Amazon SageMaker Studio in one click | Hazim Qudah | 2026-07-07 | 1,017 | -- |
| NVIDIA Isaac Teleop and GR00T 1.7 Open VLA Model Available in LeRobot | lior ben horin, Kartik S, Johnny Nuñez Cano, Edith Llontop, Leung, Andrew C Wrenn, and Shane Reetz | 2026-07-07 | 1,495 | -- |
| Atom2.7m: Representation-Level Specialization for Arithmetic-Aware Small Language Models | Maksymilian | 2026-07-07 | 2,675 | -- |
| Meet Cohere Transcribe Arabic | Shaun Cassini, Sebastian Vincent, Xiaolu Lu, Julian Mack, Dhruti Joshi, and Pierre Richemond | 2026-07-07 | 1,336 | -- |
| Why Whisper cuts off Indic transcripts after six seconds | Kavya Manohar and Kush Juvekar | 2026-07-07 | 1,452 | -- |
| Native-speed vLLM transformers modeling backend | Harry Mellor and Lysandre | 2026-07-08 | 955 | 3 |
| Data for Agents | Will Jennings, Jane Polak Scowcroft, Annie Surla, Yev Meyer, Rebecca Kao, Leanna Chraghchian, Chris Alexiuk, Michelle Xu, and Dhruv Nathawani | 2026-07-08 | 1,312 | -- |
| Distillation in 2026 (so far): which frontier models use it and how | Sergio Paniego | 2026-07-08 | 1,123 | -- |
| Putting DoctoBERT to Work: A Practical Guide | bofeng huang and Emma Scharfmann | 2026-07-09 | 3,937 | -- |
| Profiling in PyTorch (Part 3): Attention is all you profile | Aritra Roy Gosthipaty, Sergio Paniego, Sayak Paul, and Rémi Ouazan Reboul | 2026-07-10 | 4,196 | -- |
| How to visualize any Hugging Face model | Hannes von Essen | 2026-07-10 | 563 | -- |
| Can Skills Improve Codex’s Data Analysis Capabilities? | Ningyu Zhang | 2026-07-10 | 3,315 | -- |
| Can Codex Handle Real-World Data Analysis? | Ningyu Zhang | 2026-07-10 | 3,210 | -- |
| VKUE: No GPU? Runs Anyway — a 34.7B Reasoner on a Laptop … | VIDRAFT_LAB | 2026-07-12 | 1,032 | -- |
| Giving AI Agents 3D Bodies, Real Jobs, and Wallets on three.ws | three.ws | 2026-07-13 | 1,748 | -- |
| J-Space: Yet Another LLM Mind Reader? | David Louapre | 2026-07-13 | 4,714 | -- |
| Deploy GLM-5.2-FP8 as your open, frontier-level agent | Juan Julián | 2026-07-13 | 1,687 | -- |
| The state-of-the-art in open-source AI for Swiss legal tasks | Joel Niklaus and Daniel | 2026-07-14 | 2,249 | -- |
| A 0.6B model at #2 on MTEB(Law) | Stephen Solka | 2026-07-14 | 1,907 | -- |
| Welcome Inkling by Thinking Machines | ben burtenshaw, merve, Pedro Cuenca, and Aritra Roy Gosthipaty | 2026-07-15 | 3,472 | -- |
| Introducing Real World VoiceEQ: Measuring the human quality of voice AI | David Ayllon, Alice, Jeff Brooks, Franc Camps Febrer, Jakub Piotr Cłapa, Theo Lebryk, Jens Madsen, Olya Ossipova, Sharath Rao, Hoon Shin, Tigran, Rashish Tandon, and Panagiotis Tzirakis | 2026-07-15 | 1,152 | -- |
| Model Routing Is Simple. Until It Isn’t. | Yara Rizk, Eyal Shnarch, Jason Tsay, and Merve Unuvar | 2026-07-15 | 1,052 | -- |
| What building Shippy taught us about building agents | Kyle Wiggers | 2026-07-15 | 1,937 | -- |
| Security incident disclosure — July 2026 | system | 2026-07-16 | 887 | -- |
| NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB, Advancing Agentic Retrieval | Yauhen Babakhin, Ronay Ak, Jiarui Cai, Vinay Raman, Radek Osmulski, Jakub Zakrzewski, Anmol Gupta, Oliver Holworthy, Sahel Sharifymoghaddam, Khang Pham, James Rong, Steve Han, Sean Sodha, Isabel Hulseman, and Bo Liu | 2026-07-16 | 2,269 | -- |
| One Adapter, Both Modalities: Field Notes from Building and Serving a Multimodal … | Amélie Chatelain and Ishrat Jahan Ananya | 2026-07-16 | 7,302 | -- |
| Newer Models, Same Advantage | Erick Lachmann, Gabriel Pimenta de Freitas Cardoso, Francisco de Almeida Rocha Alves, and Victor Gabriel Ferreira Barbosa | 2026-07-16 | 2,359 | -- |
| When will language models be good enough? | Colin Raffel | 2026-07-16 | 812 | -- |
| Kimi K3 Model Overview: 2.8T Parameters, MXFP4 Quantization, and What the Open … | Viddi AI | 2026-07-17 | 1,046 | -- |
| Fine-tune video and image models at scale with NVIDIA NeMo Automodel and … | Pranav Prashant Thombre, linnan wang, Alexandros Koumparoulis, Wenwen Gao, Sylendran Arunagiri, and Bernard Nguyen | 2026-07-17 | 1,999 | -- |
| Aether-7B-5Attn: A 100% Open-Source Sovereign Foundation Model — and a Controlled Experiment … | VIDRAFT_LAB | 2026-07-19 | 2,299 | -- |
| Kimi K3, previewed: inside the first open 3T-class model | Hannes von Essen | 2026-07-19 | 848 | -- |
| Be Ready Before the Attack: A Practical Guide to Self-Hosting an Open … | Jeff Boudier | 2026-07-20 | 1,970 | -- |
| Introducing Cosmos 3 Edge | Pranjali Joshi and Saeed Babamohamadi | 2026-07-20 | 1,291 | -- |
| We Just Surgically Changed What Your Model Believes | Apollo Raines | 2026-07-20 | 1,091 | -- |
| Grabette: an open system to record robot-manipulation data. And build a shared … | Steve Nguyen, Claire Houziel, Gaelle Lannuzel, Simon Le Goff, Jeremy Laville, and Étienne | 2026-07-21 | 1,328 | -- |
| The State of Simulation for Physical AI: An Overview | Johnny Nuñez Cano, Mitesh Patel, Asier Arranz, lior ben horin, and Raymond Lo | 2026-07-21 | 2,044 | -- |
| Bringing Nunchaku 4-bit Diffusion Inference to Diffusers | Pham Hong Vinh and Sayak Paul | 2026-07-23 | 2,374 | -- |
| POCKET: a 35-billion-parameter model that runs on your iPhone — and on … | VIDRAFT_LAB | 2026-07-23 | 1,493 | -- |
| The influx of specialist models on the Open SLM Leaderboard | Banaxi | 2026-07-23 | 492 | -- |
| Hugging Face on AMD Instinct MI455X: First Transformers Results | Abdennacer Badaoui | 2026-07-23 | 641 | -- |
| FLUX 3 Model Overview: Multimodal Flow Models for Image, Video, Audio, and … | Viddi AI | 2026-07-24 | 1,978 | -- |
| NVIDIA Cosmos-H-Dreams: Bringing Real-Time Generative Simulation to Surgical Robotics | Lukas Zbinden, Javier Gamazo, Mostafa Toloui, and Sean Huver | 2026-07-27 | 1,712 | -- |
| FeyNoBg: A SOTA Model For Background Removal | Shreyash Nigam and Not Lain | 2026-07-27 | 1,532 | -- |
| Anatomy of a Frontier Lab Agent Intrusion: A Technical Timeline of the … | Hugo Larcher, Adrien Carreira, raphael g, and Christophe Rannou | 2026-07-27 | 6,106 | 469 |
| ECMWF's AI forecasting model is open source: now let's make it easy … | Emma Scharfmann and Harrison Cook | 2026-07-28 | 2,390 | -- |
| LFM2.5-Encoders for Fast Long-Context Inference on CPU | Fernando Fernandes Neto, Edoardo Mosca, Maxime Labonne, and Leonie Monigatti | 2026-07-28 | 1,435 | -- |
| The OlmoEarth Platform: Geospatial inference at planetary scale | Kyle Wiggers | 2026-07-28 | 1,882 | -- |
| VisionPsy-Nano: State-of-the-Art On-Device Vision-Language Models | Khurram Azeem Hashmi, Mohammadreza Zolfaghari, Changdae Park, Rishabh Jain, Nicholas Moratelli, pengfei, Louis, Tianchi Liu, Mathias Buus, and Amril Nurman | 2026-07-29 | 4,467 | -- |
| LettucePrevent - Real-Time Prevention of Factual Hallucinations in RAG | Leon | 2026-07-29 | 1,240 | -- |
| Training a 2.7B MoE from scratch for $200, one GPU at a … | V S | 2026-07-29 | 1,299 | -- |
| Bekko Embedding: how small can a multilingual retrieval model be? | Yuichi Tateno | 2026-07-29 | 2,746 | -- |
| 🎲 Apprendre à un réseau à écrire avec seulement une récompense — … | RDTvlokip | 2026-07-29 | 12,705 | -- |
| Accelerating Qwen3.6 on Intel® Core™ Ultra Series 3 with DFlash | Ofir Zafrir, Guy Boudoukh, and Igor Margulis | 2026-07-30 | 1,183 | -- |
| mDenseOn with the mLateOn: Open Multilingual, Long-Context, and Code Retrieval Models | Raphael Sourty, Antoine Chaffin, Paulo Moura, and Amélie Chatelain | 2026-07-30 | 5,088 | -- |
| GPU Management: Why Idle GPUs Are the New Grounded Aircraft | Erick Lachmann, Gabriel Pimenta de Freitas Cardoso, and Gustavo Lucchetti | 2026-07-30 | 2,604 | -- |
| The Fast Gemma Challenge: our verified-SOTA recipe, in full | VIDRAFT_LAB | 2026-08-03 | 1,058 | -- |
| 超越表层对齐：信念是通往深层对齐的新入口 | Xu Ziwen | 2026-08-03 | 1,151 | -- |
| Deploy local agents everywhere with LFM2.5-2.6B | Leonie Monigatti, Sergei Tilga, Sinoué GAD, Song Duong, Tim Seyde, and Maxime Labonne | 2026-08-04 | 1,003 | -- |
| Language Identification for 42 Indian Languages: Opening New Frontiers in Indic Speech … | Suryansh Shukla, Agneedh Basu, Sujith Pulikodan, Pranav D Bhat, and PavanKumarJ | 2026-08-04 | 1,370 | -- |
| Training a coding agent using the OpenCode harness in remote HF sandboxes … | Sergio Paniego | 2026-08-05 | 1,650 | -- |
| jina-reranker-v3.5: Faster Listwise Reranking with Hybrid Attention and Self-Distillation | Sa Zhang, Han Xiao, and Felix | 2026-08-05 | 2,018 | -- |
| IDEOGRAM-4 for inpainting with Modular Diffusers and Differential Diffusion | Alvaro Somoza | 2026-08-05 | 2,628 | -- |
| Baseten on Hugging Face Inference Providers 🔥 | Alex Ker, Roland Crosby, Sid Shanker, Johan, Célina Hanouti, Simon Brandeis, Lucain Pouget, and merve | 2026-08-06 | 865 | -- |
| Optimum Intel v2.1.0 | Ella Charlaix, Ilyas Moutawwakil, and Régis Pierrard | 2026-08-06 | 303 | -- |
| FP8 KV-Cache on Intel® Arc™ Pro B70: 2× Capacity with strong Long … | Kushal Mittal, Chun Tao, and Lokendra Uppuluri | 2026-08-06 | 3,670 | -- |
| Rebuilding Among AIs from Six Log Files | DedeProGames | 2026-08-07 | 1,945 | -- |
| TutorMoments: Do AI tutors know when to help and when to hold … | Kyle Wiggers | 2026-08-07 | 1,563 | -- |
| Lattice: an 8 MB static retriever that embeds Wikipedia in 7 minutes | Erik Kaunismäki | 2026-08-07 | 4,559 | -- |
| On the integrity of the Arabic TTS Arena leaderboard | Mohamed Rashad | 2026-08-07 | 1,272 | -- |
| Model Genome: Fingerprinting Whether an LLM Was Trained From Scratch or Derived | Proto_AGI | 2026-08-08 | 1,352 | -- |
| GLInt: Geometry-Matched Hard Negatives for Late-Interaction Retrieval | Aarush | 2026-08-08 | 4,708 | -- |
| Meta is back with Muse Glimmer: local, agentic, multimodal, and open source! | Pedro Cuenca, merve, ben burtenshaw, and Aritra Roy Gosthipaty | 2026-08-10 | 3,789 | -- |
| Making Knowledge Distillation Cheap Enough to Run at Scale | Antonio Tiene, Iker García-Ferrero, Ali Hashemi, and Bakbergen Ryskulov | 2026-08-10 | 1,621 | -- |
| FineBooks: are open OCR models good enough to unlock historical knowledge? | Sebastian Majstorovic and Daniel van Strien | 2026-08-10 | 2,151 | -- |
| Build Low-Latency Multilingual Voice Agents: Open Weights & Full Deployment Control with … | Maryam Motamedi, Mikyas Desta, Jason Li, and Jason Roche | 2026-08-10 | 1,551 | -- |
| Day-0 Muse Glimmer Support on Intel Platforms with vLLM and Hugging Face | Matrix Yao, Liangliang Ma, Jiang Li, jianan, Yi Wang, Kai Yang, Alex Gu, FanZhao, and Roger Feng | 2026-08-10 | 2,848 | -- |
| Nesso-1: Accelerating Open-Source Binding Affinity Predictions | Nikhil Shenoy, David Errington, and Francesco Di Giovanni | 2026-08-10 | 1,735 | -- |
| DAVIDAU DELETED MY POST AND BANNED ME FOR ASKING ABOUT BENCHMARKS — … | DedeProGames | 2026-08-11 | 2,594 | -- |
| Thinking of ACE? We Can Do It with Fewer Tokens | Vatche Isahagian, Jayaram Radhakrishnan, Vinod Muthusamy, Gaodan Fang, Punleuk Oum, G Thomas, Ashwath Vaithinathan Aravindan, Evelyn Duesterwald, and Merve Unuvar | 2026-08-11 | 1,636 | -- |
| Exploring NVIDIA Nemotron 3.5 Lightning: Making it see with little resources | Fabian Klemm, Thomas Endres, Henrik Klagges, Andreas Rabenstein, Lars Reimann, and Robert Dahlke | 2026-08-11 | 1,481 | -- |
| Beyond the Aggregate Score: Per-Country Domain Shift in the GWHD Wheat Head … | Saumya Saksena | 2026-08-11 | 1,427 | -- |
| Luth-2: Pushing the French Capabilities of SLMs with MOPD | Maxence Lasbordes and Guillaume Pradel | 2026-08-11 | 2,064 | -- |
| SraVaani: How Vision Helps Hearing | Agneedh Basu, Sujith Pulikodan, PavanKumarJ, Pranav D Bhat, and Suryansh Shukla | 2026-08-11 | 1,648 | -- |
| Fine-Tune SraVaani on Your Own Speech Data | Sujith Pulikodan, Agneedh Basu, PavanKumarJ, Pranav D Bhat, and Suryansh Shukla | 2026-08-11 | 2,759 | -- |
| Introducing OlmoEarth embeddings: Custom embedding exports from OlmoEarth Studio for downstream analysis | Kyle Wiggers | 2026-08-12 | 1,566 | -- |
| Meet North Micro Vision: A 2.4B Native-Resolution Vision-Language Model | David Rau and Florian Schneider | 2026-08-12 | 2,241 | -- |
| LFM2.5-VL-3B for Better and Faster Vision Capabilities for the Edge | Samuel Stevens, Ryan Shubert, Sina, Tianshu Yu, Brandon, and Leonie Monigatti | 2026-08-12 | 1,389 | -- |
| Extreme Overtraining in Tiny Language Models | Banaxi | 2026-08-12 | 832 | -- |
| Sleeper Agents and How to Tame Them | Andreas Rabenstein, Fabian Klemm, Henrik Klagges, Robert Dahlke, and Lars Reimann | 2026-08-13 | 2,112 | -- |
| What We Learned by Reproducing 2,200 papers from ICML | Abubakar Abid | 2026-08-13 | 2,009 | -- |
| Record, train, and deploy from one place with Strands Agents, LeRobot, and … | Sundar Raghavan, Steven Palma, Cagatay Cali, AWS Arron, and Yin Song | 2026-08-13 | 4,308 | -- |
| LEMUR and Mean Centering for Late-Interaction Retrieval in txtai | David Mezzetti | 2026-08-13 | 1,645 | -- |
| mLateOn: A New SoTA for Multilingual ColBERT-Style Retrieval, Evaluated on HAKARI-Bench | Yuichi Tateno | 2026-08-13 | 5,061 | -- |
| AX-Ray, Finding Causal-Leakage Defects in Two General-Purpose Public Models | VIDRAFT_LAB | 2026-08-14 | 1,937 | -- |
| State of Open Models: Summer 2026 Observations | Adina Yakefu, Apolinário from multimodal AI art, and Irene Solaiman | 2026-08-14 | 2,863 | -- |
| Who Tells You Whether the Molecule Your AI Just Designed Is Any … | VIDRAFT_LAB | 2026-08-15 | 2,975 | -- |
| Same Cluster, 33 Points More Utilization: What Changed Was the Order | Gabriel Pimenta de Freitas Cardoso, Breno de Almeida Beleza, Francisco de Almeida Rocha Alves, and Bruno Duarte | 2026-08-17 | 2,990 | -- |
| New imatrix dataset | Bartowski | 2026-08-17 | 3,722 | -- |
| Multi-Vector (Late Interaction) Embedding Models with Sentence Transformers | Tom Aarsen, Antoine Chaffin, and Raphael Sourty | 2026-08-18 | 8,611 | -- |
| How Much Memory Does Your Agent Actually Need? | Vatche Isahagian, Gaodan Fang, Jayaram Radhakrishnan, Punleuk Oum, Ashwath Vaithinathan Aravindan, Evelyn Duesterwald, G Thomas, Vinod Muthusamy, Merve Unuvar, and Ayhan Sebin | 2026-08-18 | 1,859 | -- |
| Bedrock-RL: a deterministic Minecraft framework for training and benchmarking VLM agents | Michael E | 2026-08-19 | 1,009 | -- |
| LFM2.5 Q4_0 Checkpoints from Quantization-Aware Distillation | Aditya Tadimeti and Leonie Monigatti | 2026-08-19 | 520 | -- |
| VLM Run Gateway: Run GLM-OCR, DeepSeek-OCR-2, dots.mocr with an OpenAI Compatible API | Chukwuma Nwaugha and Sudeep Pillai | 2026-08-19 | 648 | -- |
| Up to 3.2x Faster Inference with LFM2.5-DSpark | xx, Leonie Monigatti, Fernando Fernandes Neto, Tarek Dakhran, and nathan ranchin | 2026-08-20 | 1,304 | -- |
| ArmBench-ASR: A Benchmark for Armenian ASR | Alexander Shahramanyan, Mariam Avetisyan, and Hrant Davtyan | 2026-08-20 | 1,388 | -- |
| Base Optimization Stack: From Open Weights to Frontier On-Device Inference Speed | Fabian Waschkowski | 2026-08-20 | 1,338 | -- |
| Measuring benchmark optimization in speech recognition | Theo Lebryk, Eric Bezzam, Alice, David Ayllon, Jakub Piotr Cłapa, Jens Madsen, and Panagiotis Tzirakis | 2026-08-21 | 2,506 | -- |
| How Hugging Face Inference Endpoints, Jobs, and Buckets Power Search on Papers … | Niels Rogge | 2026-08-21 | 2,722 | -- |
| We changed one line and the benchmark score moved 0.21 AUROC | VIDRAFT_LAB | 2026-08-22 | 2,420 | -- |
| Writing Down the Line Between Luck and Skill | VIDRAFT_LAB | 2026-08-24 | 3,115 | -- |
| Introducing the EdgeFirst Model Zoo | Sébastien Taylor | 2026-08-24 | 1,200 | -- |
| Build Anything with gr.Workflow | yuvraj sharma and Abubakar Abid | 2026-08-25 | 1,060 | -- |
| Granite 4.2 LLMs: How They're Built | Yousaf Shah, Swanand Kadhe, Riddhiman Moulick, and Ashish Sunil Agrawal | 2026-08-25 | 4,829 | -- |
| Quantization-Aware Healing: a compressed, 4-bit model that outperforms its full-precision original | Antonio Tiene, Iker García-Ferrero, Ali Hashemi, and Bakbergen Ryskulov | 2026-08-25 | 2,084 | -- |
| Extremely Fast and Accurate Transcription with Granite Speech 5.0 Turbo CTC | Brian Kingsbury, George Saon, Samuel Thomas, Vishal Sunder, Jeff Kuo, Takashi Fukuda, Masayuki Suzuki, and Madison Lee | 2026-08-25 | 1,181 | -- |
| Training and Finetuning Multi-Vector Embedding Models with Sentence Transformers | Tom Aarsen | 2026-08-26 | 5,568 | -- |
| Modern LLMs: What the FLIP is an Engram!? | Coffee Vampire | 2026-08-26 | 1,049 | -- |
| Intel GPU AI Skills | Kushal Mittal, Zhiqi Tao, Unnikrishnan Nair, Chun Tao, Gopesh Khandelwal, Yuning Qiu, Jerry Zhang, Akash Dhamasia, Gilliean Lee, and Susan Liu | 2026-08-26 | 2,182 | -- |
| EgoSuite-Open100K: 100,000 hours of egocentric human data for Physical AI | Jonathan Stephens | 2026-08-26 | 1,199 | -- |
| TAVR: Generate Your Talking Avatar from Video Reference | Zujin | 2026-08-27 | 1,316 | -- |
| TransVLM: Detecting Any Shot Transition with Vision-Language Models | Ce Chen | 2026-08-27 | 1,358 | -- |
| Stop Abusing Git for ML Data: The Architect’s Guide to AI Storage | PRATIK PATEL | 2026-08-27 | 1,619 | -- |
| TranslatePsy-AfriSLM: Optimized Machine Translation for African Languages | Milan Gritta, Jihye Back, Patrik Lambert, Mathias Buus, and Amril Nurman | 2026-08-27 | 3,394 | -- |
| The Open ASR Leaderboard Adds Its First Global South Language | Eric Bezzam, Shobhit Banga, Manas Dhir, Bhaskar Singh, Manmeet Kaur, Aaditya Pareek, Walecha, Sagar Jain, Hanuman Sidh, and Vanshika Chhabra | 2026-08-28 | 3,002 | -- |
| Inside three.ws: The Open-Source Stack That Gives AI Agents a Body, a … | three.ws | 2026-08-29 | 4,644 | -- |
| Frontier-Assisted Single-Prompt Disposable Risk Assessment | Sonny DeSorbo | 2026-08-30 | 7,426 | -- |
| VLANeXt: A Simple and Research-Oriented Codebase for Robotics Research | Xiao-Ming Wu, Kang Liao, and Chen Change Loy | 2026-08-31 | 3,100 | -- |
| Technical writing in the agentic era | Joel Niklaus | 2026-08-31 | 691 | -- |
| From watching AI-Generated worlds to exploring them with H3 | Zuan Carreño M | 2026-08-31 | 4,155 | -- |
| Did the Civilization Emerge, or Was It Recited? | VIDRAFT_LAB | 2026-08-31 | 1,724 | -- |
| BenchMIRT: What are LLM benchmarks actually measuring? | Kyle Wiggers | 2026-09-01 | 1,527 | -- |
| Introducing @huggingface/kernels: 200+ WebGPU Kernels for Local AI | Nico Martin and Joshua | 2026-09-01 | 1,882 | -- |
| Evaluating LLMs Under Production Parity: A Replay Pipeline for Safe Model Swapping … | TechforHumans and Anna Julia Ferreira | 2026-09-01 | 3,037 | -- |
| Internet-Scale Knowledge Retrieval: A Novel Vector Search Dataset at 10B Scale | Neil Kanungo, Nathan LeRoy, and Seth Ockerman | 2026-09-01 | 1,126 | -- |
| Real-Time Intelligence with IBM Time Series Models on Confluent | Nicholas Fuller, Sean Falconer, Ayhan Sebin, Bilge Zeren Aksu, Mountu Jinwala, Pankaj Dayama, Roman Vaculin, Siddharth Bedekar, and Julian Payne | 2026-09-02 | 2,959 | -- |
| Cosmos-Reason2 on a Quest 3: Performance, Cost, and What Compression Actually Buys … | Eren Ata | 2026-09-02 | 3,556 | -- |
| Puffin-World: Scaling a Unified Multimodal Model with Native 3D World States | Kang Liao, Yihang Luo, Xiao-Ming Wu, and Chen Change Loy | 2026-09-02 | 1,855 | -- |
| TranslatePsy-Nano: Compact Multilingual Machine Translation for Resource-Constrained Edge Deployment | Kamal Gupta, Akshay Nambiar, Mathias Buus, and Amril Nurman | 2026-09-02 | 2,773 | -- |
| Training a coding model to paint watercolours with TRL and OpenEnv | Sergio Paniego | 2026-09-03 | 4,793 | -- |
| NeoMME: an efficient Multimodal-native and Multilingual Encoder | Tony Wu and Aurélien Lac | 2026-09-03 | 2,554 | -- |
| Fine-tuning a 350M Model for Better Structured Outputs in 100 GRPO Steps | Leonie Monigatti, ben burtenshaw, and Sergio Paniego | 2026-09-03 | 1,845 | -- |
| Open Yap 1K: 1,000 hours of full-duplex natural conversation, free for commercial … | Christian Vestergaard and Matias Voldby Drejer | 2026-09-03 | 1,922 | -- |
| Give Your Coding Agents a Memory You Own | David Corvoysier | 2026-09-03 | 1,645 | -- |
| From Zero to Hero: An Open LLM Ecosystem for Armenian | Erik Arakelyan, Vahan Martirosyan, Sergoyan, Heghine Grigoryan, Nane Khachatryan, Mery Davtyan, Khatun Avetisyan, and Hayk Shahsuvaryan | 2026-09-04 | 2,386 | -- |
| VLM Run Gateway: Run open-weight OCR, VLM and vision models behind one … | Sudeep Pillai and vlmrun | 2026-09-04 | 507 | -- |
| TurboQuant Quantization Explained | Tarun Jain | 2026-09-06 | 1,735 | -- |
| A Real-World Dataset for Noise-Robust Speech AI: 100+ Timestamped Noise Types Across … | Suryansh Shukla, PavanKumarJ, Sujith Pulikodan, Agneedh Basu, Pranav D Bhat, and Prasanta Kumar Ghosh | 2026-09-07 | 1,184 | -- |
| Making open-source AI weather forecasting models easy to run | Emma Scharfmann, Aaron Spring, Ryan Abernathey, and Kashif Rasul | 2026-09-08 | 3,856 | -- |
| Safety for Whom? Refusing the Right Subset of a Topic, Not the … | Antonio Tiene, Alejo Lopez Avila, and Iker García-Ferrero | 2026-09-08 | 1,437 | -- |
| So, let’s make our own dataset | tegridydev | 2026-09-09 | 5,936 | -- |
| IBM releases SOTA Granite Time Series PatchTST-FM-r2 model with commercial-friendly license | Roman Vaculin, Wesley M. Gifford, Jiri Navratil, Chandra Reddy, and Ayhan Sebin | 2026-09-09 | 1,701 | -- |
| Was sind KI-Agenten? So funktioniert Agentic AI | K | 2026-09-09 | 1,837 | -- |
| Trained 210M text-to-image model from scratch on one GPU: what actually mattered | Ivan Mikhnenkov | 2026-09-09 | 1,731 | -- |
| Per-tensor layout maps for GGUF quantization | Bartowski | 2026-09-10 | 2,631 | -- |
| Rebuilding AUTOMATIC1111 with Gradio Workflow | yuvraj sharma and Abubakar Abid | 2026-09-10 | 2,025 | -- |
| Lorem Ipsum | The Supreme GGUF-SafeTensors Guy | 2026-09-10 | 522 | -- |
| Async GRPO with LoRA across HF Jobs: a bucket, a proxy, and … | Amine Dirhoussi, Quentin Gallouédec, Kashif Rasul, and Sergio Paniego | 2026-09-10 | 6,317 | -- |
| One sandbox per rollout, or how labs run RL for agents in … | Sergio Paniego | 2026-09-11 | 2,953 | -- |
| How Did We Get Here? | Nilky | 2026-09-13 | 525 | -- |
| I love speed. We all love speed. But what happens when the … | Sonny DeSorbo | 2026-09-13 | 4,155 | -- |
| The Hugging Face Model That Nobody Asked For | The Supreme GGUF-SafeTensors Guy | 2026-09-13 | 640 | -- |
| Same bytes, closer to the original: two lines of AutoRound we had … | VIDRAFT_LAB | 2026-09-14 | 1,856 | -- |
| 适配器即模型 —— ShadowPEFT 现已集成到 🤗 PEFT 库 | Zx Li, SeanLee, HSURA, and Jing Li | 2026-09-15 | 1,036 | -- |
| Your Inference Server is Secretly a Learner: Reef Infrastructure for Continual Self-Improving … | Ao Qu, Bo Liu, Han Zheng, Zhou Zijian, Xuan Jiang, Simon Yu, zoedusy, kaichen-z, XINKAI ZOU, Wenhao Chai, Chonghe Jiang, and hao he | 2026-09-15 | 1,929 | -- |
| What If the Adaptation Were a Model? ShadowPEFT in 🤗 PEFT library | Zx Li, SeanLee, HSURA, and Jing Li | 2026-09-15 | 2,631 | -- |
| Your Agent Aced the Task. Will It Do It Again? | Evelyn Duesterwald, Lilian Ngweta, Vatche Isahagian, Jayaram Radhakrishnan, Vinod Muthusamy, Gaodan Fang, Ashwath Vaithinathan Aravindan, Punleuk Oum, G Thomas, Merve Unuvar, Ayhan Sebin, and Michał Ulewicz | 2026-09-15 | 2,031 | -- |
| OpenRouter Leaderboard: 425 Models by Price, Measured Speed and Korean Quality | ginigen-ai | 2026-09-16 | 1,360 | -- |
| funes: Local Memory for Coding Agents, Built on Lance | Aritra Roy Gosthipaty and Ayush Chaurasia | 2026-09-17 | 805 | -- |
| When Benchmark Numbers Become Marketing | DedeProGames | 2026-09-17 | 4,840 | -- |
| BananaMind 2 Pro: We've (almost) matched SmolLM2 at 20x fewer tokens... Trained … | Banaxi | 2026-09-18 | 1,568 | -- |
| Optimum-Intel v2.2.0 & OpenVINO GenAI 2026.4.0: What's New | Ella Charlaix, Anastasiia Pnevskaia, Altana Natyrova, Artur Paniukov, and Roman Kazantsev | 2026-09-18 | 872 | -- |
| Layer-Feedback Transformer (LFT) | Banaxi | 2026-09-19 | 1,755 | -- |
| Introducing jev-reranker: Reranking and Relevance Filtering for RAG | Yuichi Tateno | 2026-09-19 | 1,435 | -- |
| Your model already knows it's wrong. Asking costs 0.06 seconds and zero … | VIDRAFT_LAB | 2026-09-20 | 924 | -- |
| Inside the JEV Ecosystem: 13 Answer Verifiers on One Test Set | Proto_AGI | 2026-09-20 | 1,416 | -- |
| Pruning LLMs Like a Physicist: Block Removal as an Ising Optimization Problem | Antonio Tiene, Ali Hashemi, David Jansen, and Roman Rausch | 2026-09-21 | 1,877 | -- |
| relore - repository memory for coding agents | Tarek Ziadé | 2026-09-21 | 991 | -- |
| What Is Jev AI? A Practical Guide to System One and Executable … | bna | 2026-09-21 | 2,662 | -- |
| tokenizers v1: encode, decode and scaling, measured | Arthur Zucker, Simon Brandeis, Luc Georges, and Lysandre | 2026-09-21 | 2,333 | -- |
| Jev AI vs LLMs: When Should You Use a Decision Model Instead … | bna | 2026-09-21 | 1,843 | -- |
| How UK AISI and EvalEval Are Making Benchmark Results Reproducible | Avijit Ghosh, Jenny Chim, Deep Joshi, Srishti, Matt Kennedy, Irene Solaiman, Jessica McFadyen, Lynn Tan, and Coz | 2026-09-22 | 893 | -- |
| Compute:Arena / Measuring Local Inference Across Models, Quants, Chips, and Runtimes | Fabian Waschkowski and Lukas Wesemann | 2026-09-22 | 639 | -- |
| Jun Kim, oMLX creator and maintainer, joins Hugging Face to support the … | Pedro Cuenca, Lysandre, Victor Mustar, Julien Chaumond, and Jun Kim | 2026-09-22 | 702 | -- |
| How to Use the Jev AI Model: A Step-by-Step Developer Guide | bna | 2026-09-22 | 1,838 | -- |
| Transformers now runs llama.cpp quants | Marc Sun, Arthur Zucker, and Lysandre | 2026-09-22 | 2,443 | -- |
| Jev ai vs djev vs Laya vs OpenJev vs SemIf: Which Decision … | bna | 2026-09-23 | 2,150 | -- |
| Know Who Spoke When: Build Real-Time, Multi-Speaker AI with NVIDIA Nemotron 3 … | Francesco, Ivan Medennikov, Taejin Park, Tatiana Timofeeva, Jagadeesh Balam, Adi- margolin, and Maryam Motamedi | 2026-09-23 | 3,324 | -- |
| How to Use NVIDIA Warp and MjWarp to Accelerate Robotics Simulation and … | Johnny Nuñez Cano, Asier Arranz, Rishabh Chadha, and Ben Oliveri | 2026-09-23 | 3,515 | -- |
| Basis Conversations 1500: 1,500 hours of multilingual, multi-party conversation | Alexis Sursock | 2026-09-23 | 779 | -- |
| FLUX 3 Action: a world action model you can fine-tune | Sumith Kulal, Patrick Esser, A S, Stephen Batifol, and Maanav D | 2026-09-23 | 1,426 | -- |
| The Road to Open Source Hell Is Paved With Good Intentions | Quentin Gallouédec | 2026-09-23 | 1,197 | -- |
| Accelerating vision-language models with LFM2.5-VL-DSpark | xx, Yuri Khrustalev, Leonie Monigatti, and Viviana Márquez | 2026-09-24 | 1,011 | -- |
| How to Train Your Robot: The LanceDB Edition | Caroline Pascal, Ayush Chaurasia, Taylor Smith, and Quentin Lhoest | 2026-09-24 | 2,600 | -- |
| AI Agents List [2026] | Frameworks, Agentic Harnesses & Useful Repos | tegridydev | 2026-09-24 | 3,405 | -- |
| Jev vs Laya: Hosted API or Open Weights? (2026 Guide) | bna | 2026-09-24 | 1,020 | -- |
| From Agent-Based Models to LLM Agents: How Artificial Societies Are Changing | Anouar Imel | 2026-09-24 | 2,696 | -- |
| Bringing Humanoids to LeRobot | Martino Russi, Steven Palma, Pepijn Kooijmans, Khalil Meftah, Maxime Ellerbach, Nikodem Bartnik, Nicolas Rabault, Caroline Pascal, Etienne Chassaing, Thomas Wolf, Lysandre, and Kartik S | 2026-09-25 | 1,934 | -- |
| The DGX Spark Handbook | 0xSero | 2026-09-25 | 5,954 | -- |
| Pre-training a 1.11B LLM on a 6 GB Laptop GPU — Measured, … | Rambarun Komaljeet | 2026-09-26 | 3,253 | -- |
| Laya AI Model: How It Works, Run It Locally, and Evaluate It | bna | 2026-09-26 | 2,961 | -- |
| YODAS v3: A 1 Million Hour Dataset for the Next Generation of … | William Chen, Shinnosuke Takamichi, Sayaka Shiota, Satoru Fukayama, and Shinji Watanabe | 2026-09-27 | 1,307 | -- |
| autotrust/JEV-27B: fast, calibrated decisions and full reasoning from one open model | Josh Liu, Hai Yu, Daniel Tang, ajing, and Jeff | 2026-09-27 | 2,630 | -- |
| Holo4: powering generalist computer-use agents | Tony Wu, Maxime Theillard, Frederic Renard, Vincent Coyette, Emrick Sinitambirivoutin, Avshalom Manevich, Antonio Loison, Antoine Bonnet, Maxime Langevin, Aleix Cambray (H-AI), Léonard Benedetti, Tony Wu, Mats L. Richter, Michael Eickenberg, Sławek Mucha, | 2026-09-28 | 1,616 | -- |
| Robotics benchmarking: where we are, and what comes next | Luca Cilio | 2026-09-28 | 2,508 | -- |
| Open Superconductor Challenge: Help Discover the Next 2D Superconductor — From Your … | VIDRAFT_LAB | 2026-09-29 | 956 | -- |
| Getting the Source Right, Not Just the Fact: Source-Aware Verification for MCP … | Antonio Tiene, Ander Alvarez Sanz, Oliver Wirjadi, and Alessandro Genuardi | 2026-09-29 | 1,977 | -- |
| NVIDIA Kumo Tabular Sets a New Accuracy-Efficiency Frontier for Tabular Prediction | Jingang Qu, Valter Hudovernik, Martin Jurkovic, Cedric Lorenz, Akihiro Nitta, Dmitry Gordeev, Federico Lopez, Ramona Bendias, Gilberto Titericz Jr, Aleksandar S. Sokolovski, Isabel Hulseman, Jure Leskovec, and Matthias Fey | 2026-09-29 | 1,738 | -- |
| Anatomy of a bug-fixing agent | Tarek Ziadé | 2026-09-29 | 2,427 | -- |
| Qwen3.8-27B-pi: Effort-Ordered Reasoning for Agentic Coding | Thomas Kim | 2026-09-30 | 7,028 | -- |
| What Is OpenAI Decisions API? A Practical Guide | bna | 2026-09-30 | 2,829 | -- |
| Open TTS Leaderboard: Scalable Evaluation for Multilingual Text-to-Speech and Voice Cloning | Eric Bezzam, Steven Zheng, Eustache Le Bihan, and mrfakename | 2026-09-30 | 1,674 | -- |
| autotrust/JEV-27B-VL: a decision model that learned to see without a single image … | Hai Yu, Josh Liu, Daniel Tang, Jeff, and ajing | 2026-09-30 | 1,073 | -- |
| Introducing Olmo-core 3: Open, scalable training infrastructure for large MoEs | Kyle Wiggers | 2026-10-01 | 1,437 | -- |
| AutoSynthData: Generating Training Data for Enterprise Agents | Esakkivel Esakkiraja, Shruthan Radhakrishna, Denis Akhiyarov, and Sagar Davasam | 2026-10-02 | 2,242 | -- |
| Open-sourcing AstaBrief, the fast report-generation model in Asta | Kyle Wiggers | 2026-10-02 | 2,815 | -- |
| New in llama.cpp: Decision Models | Xuan-Son Nguyen and Victor Mustar | 2026-10-02 | 1,021 | -- |
| Data-center AI, now on a laptop — POCKET-Darwin-180B | VIDRAFT_LAB | 2026-10-02 | 1,058 | -- |
| Live Human Feedback in the Training Loop: Aligning Diffusion Models with Real … | Mads Kuhlmann-Joergensen | 2026-10-02 | 2,210 | -- |
| The Agent Said It Was Done. The Database Disagreed. | Tuhin Kundu | 2026-10-03 | 3,146 | -- |