Hugging Face
Hugging Face is one of the entities PulseAugur tracks across the AI industry. This page surfaces every recent cluster mentioning Hugging Face — vendor announcements, third-party press, social commentary, research papers, and regulatory filings — ranked by signal across our 200+ source set. Linked to the canonical entity record on Wikipedia and Wikidata so the entity card AI engines build is grounded in the same identity Wikipedia uses, not a slug-collision lookalike.
- developed by transformers 95%
- instance of Apache Software License 2.0 95%
- developed by SenseNova U1 95%
- instance of SenseNova U1 95%
- acquired Gradio 95%
- used by Gemma 4 90%
- used by DeepSeek V4 90%
- partners with NVIDIA 90%
- developed Transformer Reinforcement Learning 90%
- used by Lens 90%
- developed by Lens 90%
- partners with Intel 90%
- 2026-06-09 product_launch Hugging Face launched an integration allowing GitHub Actions to run on Hugging Face Jobs infrastructure. source
- 2026-05-19 product_launch Hugging Face released new tools and features for Gradio, enabling custom AI front-ends and introducing the Ettin Relinker. source
- 2026-05-18 product_launch Hugging Face launched the Open Agent Leaderboard.
- 2026-05-14 controversy Infostealer malware campaign discovered on Hugging Face, impersonating OpenAI. source
- 2026-05-14 research_milestone Hugging Face published a blog post detailing a new method for asynchronous batching to improve LLM inference performance. source
- 2026-05-11 product_launch Hugging Face launched three new open-source projects: Daggr, custom CUDA kernels, and OpenClaw. source
- 2026-05-10 controversy A fake OpenAI repository impersonating a privacy filter model distributed malware on Hugging Face.
30 day(s) with sentiment data
-
Total Commander plugin maps Hugging Face repos as virtual file system
A new plugin for Total Commander allows users to access Hugging Face repositories as a virtual file system. This tool enables direct browsing, viewing file details, and downloading content from Hugging Face directly wit…
-
Google releases Gemma 4 checkpoints for efficient AI training
Google has released Gemma 4 checkpoints optimized for quantization-aware training (QAT) on Hugging Face. This advancement allows for more efficient model deployment, a feat that was not possible on a single consumer GPU…
-
Run Google's Gemma-4 12B model on WSL2 with llama.cpp
A guide details how to run Google's Gemma-4 12B model on Windows Subsystem for Linux 2 (WSL2) using the llama.cpp framework. The process involves updating the WSL environment, installing necessary dependencies like buil…
-
Google releases Gemma 4 QAT checkpoints for faster on-device AI
Google has released quantization-aware training (QAT) checkpoints for its Gemma 4 models, significantly reducing their memory footprint and increasing inference speed on consumer hardware. These new checkpoints allow fo…
-
iOS app GenBench enables on-device GGUF model benchmarking
A new free iOS application called GenBench has been released, allowing users to download, run, and benchmark GGUF models directly on their iPhones and iPads. The app utilizes llama.cpp and Metal for offline operation an…
-
Cohere joins Hugging Face webinar on ASR advancements
Cohere is participating in a Hugging Face webinar discussing advancements in automatic speech recognition (ASR). The webinar will feature insights from CMU's Shinji Watanabe and Cohere's Julian Mack, focusing on the new…
-
Hugging Face Hub adds storage buckets for AI datasets
Hugging Face has introduced storage buckets to its Hub platform, offering a new way to manage and store large datasets and models. This feature aims to streamline the workflow for AI developers by providing dedicated, s…
-
OBLITERATUS/Gemma-4-12B-OBLITERATED model available on Hugging Face
The OBLITERATUS/Gemma-4-12B-OBLITERATED model is now available on Hugging Face, offering users detailed instructions for integration with various libraries and tools. These include popular frameworks like Transformers a…
-
Step 3.7 Flash leads benchmarks in speed, cost, and performance
StepFun's new model, Step 3.7 Flash, has achieved top rankings on the Artificial Analysis (AA) benchmark, excelling in speed, cost-efficiency, and end-to-end performance. The model demonstrates impressive output speeds …
-
Developer releases MIT-licensed tool for exploring Kokoro AI model
A developer has created a tool to explore the Kokoro model, making the code MIT licensed. The project includes a specific Kokoro code repository and trained models available on Hugging Face. Pre-compiled Windows version…
-
Comfy-Flow.com adds model linking and search improvements
The website Comfy-Flow.com has been updated with enhanced search filtering and additional tags. A new feature displays a "Required Models" card for each workflow, listing the models used and providing links to their Civ…
-
New tool optimizes llama.cpp models with advanced NVFP4/MXFP6 quantization
A developer has released an advanced quantizer tool for llama.cpp, designed to create NVFP4 and MXFP6 GGUF models. This tool goes beyond basic quantization by evaluating various methods and incorporating custom techniqu…
-
MOSS-TTS-v1.5 Tops Hugging Face TTS Trending with 20.6K Downloads
MOSS-TTS-v1.5, a text-to-speech model, has achieved the top position on Hugging Face's trending list for its category. The model has garnered significant attention, evidenced by over 20,600 downloads. Further details ab…
-
Hugging Face releases AI updates for LeRobot, Ulysses, and RL training
Hugging Face has released updates across several AI projects. LeRobot v0.5.0 introduces scaling across all dimensions, while Ulysses implements sequence parallelism for training with a 1 million token context window. Ad…
-
JoyAI-Echo video model released for long-form content
A new video generation model called JoyAI-Echo has been released on Hugging Face, built upon the LTX-2 architecture. This model is designed for creating long-form video content, featuring capabilities such as minute-lev…
-
Cohere releases North-Mini-Code-1.0 coding model
Cohere has released North-Mini-Code-1.0, a 30 billion parameter coding model. While its general artificial analysis score is lower than some competitors, it performs competitively in coding benchmarks. The model is avai…
-
Cohere Labs releases BLS-Mini-Code-1.0 model on Hugging Face
Cohere Labs has released BLS-Mini-Code-1.0, a new model available on Hugging Face. The model has seen over 1,700 downloads in the past month. Currently, there is no model card available to provide further details on its…
-
New 2B-parameter TTS model dots.tts achieves SOTA
Researchers have introduced dots.tts, a 2 billion parameter text-to-speech model that operates in a continuous latent space. The model incorporates several innovations, including an AudioVAE for a structured speech repr…
-
UniSHARP enables universal monocular view synthesis across diverse cameras
Researchers have developed UniSHARP, a novel method for photorealistic view synthesis that extends the capabilities of the existing SHARP technique. UniSHARP can render high-quality novel views from a single image captu…
-
PaperFlow framework enhances scientific paper recommendations
Researchers have developed PaperFlow, a new framework designed to improve scientific paper recommendations by mimicking the dynamic nature of daily reading habits. Unlike static ranking systems, PaperFlow dynamically pr…