OPEN SOURCE = UNLIMITED SOVEREIGN INTELLIGENCE
Open source AI models and their underlying code enable organizations and nations to develop artificial intelligence capabilities independently, without reliance on proprietary systems or external vend
Knowledge catalogue
Open source AI models and their underlying code enable organizations and nations to develop artificial intelligence capabilities independently, without reliance on proprietary systems or external vend
.@Qualcomm is expanding its collaboration with @huggingface to scale open, developer-driven AI. From model onboarding to agentic workflows across edge and data center, this simplifies how developers b
Rampart from the @ndstudio @WhiteHouse is number one trending token classification model on HF. Very cool to see public organizations starting to own and build their weights instead of renting them fr
Rampart, our PII removal model, has cracked the first screen of the top trending models across any category on Huggingface, on the same tier as GLM 5.2 / Deepseek! If building systems at fast pace at
Clem Delangue argues that AI companies should not be referred to as 'labs,' characterizing such terminology as pretentious and inaccurate since these organizations operate as commercial enterprises ra
Clem Delangue announced the term 'Summer of Open-source AI' in a live discussion with Dee Bosa and Vipul Ved, indicating a period focused on advancing and promoting open-source artificial intelligence
why are the open tools so low in the list? We need to improve integration between open platforms and open models @steipete @thdxr @Teknium @badlogicgames! Coding agents are real users of the @huggingf
“Agentic kernel optimization is the future of on-device inference” @xenovacom used Fable 5 to write kernels that pushed Gemma 4 to a massive 255 tok/s on WebGPU with M4. He shared the demo, so you can
AI systems that are accessible or even better open-source are safer for a simple reason: more people can inspect them, test them, stress them, and report what breaks or harms to keep the builders acco
GLM 5.2 just became the first open-source model to lead a category on APEX-SWE. It scored a 55.3% Pass@1 on Integration, the top score we've recorded for any model, open or closed source. On the overa
Pivot 0.5 is the first trained model released by the organization, available as a free research preview with open weights coming soon to Hugging Face. The announcement was made by Clem Delangue on X (
Portugal has launched its first open-source AI model as part of Europe's broader initiative to achieve 'tech sovereignty' and reduce dependence on non-European AI systems. This development aligns with
This is absolutely correct. The only way to fight 'open' is with 'open.' Glad to hear a voice in Washington that understands the real stakes. 'At this very moment China is giving its AI technology awa
A study from @Stanford showed that 71.3% of chatgpt queries could be accurately answered by a local model. I suspect a major part of enterprise AI workloads could be run locally too for free (compared
'At this very moment China is giving its AI technology away. It's releasing open-weight AI models that are cheap, capable, and they're fast becoming the world's default.' We can overcome this. @neil_c
GLM-5.2 has become the most-liked model of all time on Hugging Face's model repository, surpassing previous records. This achievement reflects significant community recognition and adoption of the mod
if US bans open source because China is leading, then 1. impossible to enforce(you just need to download a zip for weights) 2. Increases the cost of using AI for US companies and consumers, while Chin
NEW on Hugging Face: Hardware filters 🖥️ A new Hardware filter on the Models page results to models that fit a specific GPU, CPU, or Apple Silicon chip, so you only see what will actually run on your
Ornith-1.0-35B, a 35-billion parameter model, has been made available for use through Claude Code via Hugging Face integration. This announcement indicates expanded model availability and integration
Super excited about open-source router systems and routing models like @vllm_project semantic router: https://huggingface.co/llm-semantic-router The future is multi-models and you'll want to customize
This post by Clem Delangue discusses the current state of the timeline regarding closed versus open source approaches, arguing that open source should prevail. The tweet appears to address the competi
The food delivery leader in China just dropped an open weights 1.6 trillion parameter model while half the USA was sleeping … 🫨 🇨🇳 Meituan, China's largest food delivery platform, open-sourced a 1.6 t
𝗚𝗟𝗠-𝟱.𝟮 (the latest open weights model) is having an Enterprise moment, and it is not an exaggeration.🚀 🔥 We have been impressed by how strongly GLM-5.2 is pushing long-horizon performance .. not just
We are making a deliberate effort to use GLM 5.2 with OpenCode internally at Jarvislabs. I have spoken to 3 enterprise customers last week, who are exploring to host multiple open source models and mo
You can embed this model practically anywhere - like a chrome extension This is transformersjs + rampart for real-time PII removal in your browser (personal email still blurred) You can see I'm toggli
You cannot ban Chinese open source to make US win. The whole point of open source is it's OPEN. You might ban in the US, sure, but how are you going to ban it outside US? Instead the real question is,
You wake up. It’s 2026. The newest open source AI model was trained by federal government employees, Big Balls and Groot. FULL INTERVIEW: Engineers Edward Coristine (@as400495) and Tai Groot (@taigrr)
🚨 A year ago, OpenMed didn't exist. Today: 340M model downloads. 1,500+ open medical models, all Apache 2.0. 650+ run on your iPhone, fully on-device. Started by one person. Now built by a community.
The Hugging Face CEO argues that being characterized as 'too dangerous' can serve as effective marketing for frontier AI companies, suggesting that such controversy may enhance their public profile an
FULL INTERVIEW: Engineers Edward Coristine (@as400495) and Tai Groot (@taigrr) just released an ML model called Rampart for the National Design Studio. It's a local-first, open-source AI privacy model
RAMPART is a model or tool from National Design Studio available on HuggingFace that appears to focus on design-related applications or generative capabilities. The project was shared by Clem Delangue
I just published my article on how hyperscalers like AMZN, MSFT, and $GOOGL will benefit in the token optimization era and why I believe the companies will re-rate much higher. https://www.uncoveralph
Clem Delangue argues that open-source AI development offers significant benefits for technological progress, market competition, and system transparency while presenting substantially lower risks comp
Open source AI is gaining traction due to the White House’s clampdown on Anthropic & OpenAI. “A lot of founders and developers have been burned by trying to build apps on top of these really advanced
SITUATION EXPLAINED: What even is a distillation attack? @DanielleFong, founder of LightCell Energy: 'I understand you don't wanna get automatically hit by a bunch of subscriptions and then distilled
SITUATION EXPLAINED: Why is restricting open source AI models structurally impossible? @ClementDelangue, co-founder and CEO of @huggingface: 'Open weights are fundamentally different than an API. The
The U.S. Government is launching machine learning models on Hugging Face, a popular open-source platform for sharing AI models and datasets. This initiative, highlighted by Hugging Face co-founder Cle
THIS IS WHY WE MUST SUPPORT THE OPEN SOURCE APP LAYER MORE OPEN SOURCE AI = INDIVIDUAL SOVEREIGNTY Remarkable that some people fondly miss Googles RSS reader app, already gone for over a decade. Remar
Tired: The US government regulating open-source AI models Wired: The US government training and releasing open-source AI models Today, we are releasing Rampart: a 14.7MB machine learning model designe
Today, we are releasing Rampart: a 14.7MB machine learning model designed to protect citizens’ privacy by redacting personal information directly in your browser before it gets sent to any server A bi
Tomorrow, we're releasing an ML model for on-device PII redaction, which is far smaller than existing models, but is surprisingly effective. It is a first iteration. Will be live at 12:30! A big relea
We quantized GLM-5.2 (744B MoE) to 4-bit — and kept its MTP draft head in BF16. → Matches the FP8 release on quality → Runs on 4×H200 instead of 8 → Fastest 4-bit GLM-5.2 at int conc: +69–79% vs AWQ /
Britain didn't industrialize first by forcing its companies to use more expensive coal. Why would we force American companies to use more expensive intelligence? AI is rapidly becoming a foundational
Clem Delangue suggests that government regulation of AI models perceived as 'dangerous' can paradoxically serve as effective marketing, particularly for enterprise sales. The implication is that regul
Good guy DeepSeek gives us accelerated models The most interesting one here is Gemma4-12B, I presume vision included. Might be the best local model in its weight class now, by some margin Qwen 3.5 not
https://huggingface.co/collections/deepseek-ai/deepspec Good guy DeepSeek gives us accelerated models The most interesting one here is Gemma4-12B, I presume vision included. Might be the best local mo
Llama.cpp has merged DFLASH support into its main branch, which is a development update related to optimization or acceleration technology for running large language models locally. The post expresses
omg it's basically pi, but more minimal for educational purposes and in python. @marlene_zw is gonna love this. great educational resource, alejandro! introducing tau τ — an educational agent harness
the future of AI is multi-model (including a majority of open-source ones provided by @huggingface of course!)! How to keep AI spend flat while token usage grows exponentially: Not with friction and s
America is about to lose the AI race, and it will not happen at the frontier. It will happen at the floor. Everyone is watching who ships the smartest model. The actual war is over the 80% of tokens n
Every enterprise will have its own model-harness-sandbox-eval flywheel with token value per watt optimization. This is the future. Simple reason: tacit knowledge about the domain and customers and the
How to keep AI spend flat while token usage grows exponentially: Not with friction and spend alerts. With better defaults, routing, and caching. Better Defaults (not Usage Caps) – Engineers can choose
not dunking on Dario here bc Mythos is to GPT2 what a lion is to a bumblebee, but it’s insane that GPT-2, and successive models, have had ~0% impact on causing bad universal outcomes & so SHOULD HAVE
VCs are now sharing screenshots in group chats of Claude discouraging investment in open-source AI infra startups and models. Obviously there is an absolute EXPLOSION of pitches in inference companies
Fun fact: Dario once delayed the release of GPT-2 back at OpenAI, claiming it was too dangerous Dario fearmogged so hard that global AI progress got halted They could have quietly released it as Opus
NVIDIA's GLM-5.2-NVFP4 is a quantized version of a large language model optimized for inference efficiency using NVIDIA's proprietary quantization format. The model is hosted on Hugging Face and repre
I cannot verify the claims in this post as the URL format appears invalid and the specific version numbers (GLM 5.2, Claude Code/Opus 4.8) don't correspond to publicly documented model releases as of
If you are on the verge of AGI or ASI, why isn’t your model smart enough to recognize espionage distillation in real time? You say “cure cancer in a few years.” Isn’t sniffing illicit distillation qui
Clem Delangue, CEO of Hugging Face, posted a brief announcement indicating his imminent participation in an event or discussion. Without access to the full context, this appears to be a casual notific
This post discusses apparent contradictions in claims about Mythos' cybersecurity capabilities, suggesting tension between assertions that it excels at cyber defense versus reports that it failed to d