Knowledge catalogue
I'm pretty sure it can be done with any model based on qwen4exp, which Qwen's next local models will be based on. You can use a quant that barely fits in VRAM and still run at the model's maximum cont
arXiv:2609.15759v1 Announce Type: new Abstract: Full-duplex speech-to-speech (S2S) models enable natural conversational AI by allowing simultaneous listening and speaking. However, these models typica
THE SANDBOX WAS A PROP! OpenAI Turned Off the Guardrails, Left a Door to the Internet, and Then Sold the Hugging Face Breach as “Rogue AI” It is time to understand how you were lied to and by whom. In
arXiv:2609.12993v1 Announce Type: cross Abstract: We investigate the effect of pretrained T5 model scale and explicit motion features on pose-to-text Indian Sign Language Translation (SLT) for the WSL
Hi everybody, we post-trained Qwen 3.8 27B to be more efficient by figuring out which tokens were linked to overthinking and penalizing them without 'attacking' the reasoning length directly then fixe
The first generation of our 150M model has just been released Its performance is similar to that of GPT2-Small The benchmarks: PIQA: 62.24% Hellaswag: 32.20% Arc-Easy: 44.91% Arc-Challenge: 25.00% Ari
Automated lint: 12 errors, 54839 warnings, 3 info
arXiv:2509.22360v2 Announce Type: replace Abstract: Large language models (LLMs) excel at operating at scale by leveraging social media and various data crawled from the web. Whereas existing corpora
People keep on getting confused about this, so I looked at the safetensors on hf. To be precise, the main model about 551.566B parameters with 40 layers. The FFN experts total to 543.582B parameters,
After noticing that it is ranked among MUCH larger frontier models in the EQ-Bench Creative Writing benchmark and the Hemingway-bench, I decided to give it a try and was very impressed. I didn't do ve
arXiv:2609.05881v1 Announce Type: cross Abstract: Developers increasingly run large language models locally by pulling quantized GGUF artifacts from public registries, yet nothing in the distribution
(hopefully this is okay to here - it seems like audio models and image / video modeals is allowed but yeah this is a bit different) So I've been doing independent audio research for a while now. The u
In my latest op-ed for @TIME, I discuss why the OpenAI Hugging Face cyber incident represents a turning point for AI safety. If we want to prevent more autonomous cyberattacks from threatening critica
Automated lint: 12 errors, 53990 warnings, 3 info
Technology is only good if it empowers you. If it locks you down, saps your attention, or steals your ideas, then it should be discarded. LLMs could go either way. That's why we are dedicated to open,
NVIDIA introduced two Rust projects for GPU development in September 2026: `cuda‑oxide`, a nightly‑only backend that compiles SIMT‑style kernels to PTX via LLVM and uses DisjointSlice and launch contr
This weekend I posted about the gap closing between frontier models and open source models. Well, now I'm coming with receipts. I've been running local + cloud models against real public github codeba
Sometimes I wonder what would have happened if we hadn't disclosed the agent cyberattack publicly. It reinforced my conviction that we need 100x more transparency in AI. Otherwise we're going to be in
arXiv:2609.05241v1 Announce Type: new Abstract: A rapidly expanding ecosystem of actors is removing built-in safety guardrails from open-weight AI models. We profile this ecosystem by identifying key
I've been running Qwen3.8-27B as a local inference server for a production content intelligence pipeline (HVAC industry stuff, lots of long-context retrieval and structured extraction). I have been wa
M.G. Siegler / Spyglass: By declaring that GPT-6 Astra has ushered in the AGI era, OpenAI is being flippant and cementing the term's status as nothing more than marketing — With ‘GPT-6 Astra’, AGI is
Moving Large Language Models (LLMs) from experimental prototypes into enterprise production exposes a critical truth: your infrastructure dictates both your performance ceilings and your unit economic
the united states split the atom, landed on the moon, beat polio, doubled the global wheat yield, sequenced the genome, and put a computer in every pocket. we've taken steps backwards too, like doom s
Congrats @ClementDelangue @Thom_Wolf!!!!!!!! Big day for sovereign ai. excited to see open weight models, open science, (and most importantly, micro ducks) find their forever home 🤗🦆 Exciting day for
Hugging Face is in great hands. Big win for open source. And do not underestimate NVIDIA in its open-source efforts. They have been shipping great open models, and I think this amplifies their efforts
I liked the Nvidia that focused on just GPUs for gaming, not on the Nvidia of today which seem want power consolidation. Modelscope is another platform for those that simply want to know an alternativ
arXiv:2603.16749v3 Announce Type: replace Abstract: Large language models (LLMs) are increasingly deployed in applications with societal impact, raising concerns about the cultural biases they encode.
Without Hugging Face the launch and scaling of stable diffusion & the models that followed would only have been a fraction of what they were. Great and super engaged team who have constantly flown the
A privacy gate on your Mac detects PII before it leaves your device. Choose to run that step locally, send the file to the cloud, or skip entirely. We are open-sourcing the classifier today. Read the
I hadn’t realized it until now, but could Microduck actually be the biggest launch of a new consumer robot ever? 10,500 robots ordered and $4.54M in sales in just 4 days. It’s surprisingly hard to fin
“increasing agent autonomy can gradually make human oversight ineffective by causing approval fatigue, overreliance, loss of situational awareness, and skill degradation.” new paper from @mmitchell_ai
arXiv:2608.30114v1 Announce Type: new Abstract: Brazilian Portuguese remains under-served by open language models, and the few that exist are difficult to reproduce and are often compared without meas
arXiv:2603.00620v3 Announce Type: replace Abstract: Multilingual NLP research increasingly involves hundreds or thousands of languages across different datasets. Managing, discovering, and reporting l
seems like it I hadn’t realized it until now, but could Microduck actually be the biggest launch of a new consumer robot ever? 10,500 robots ordered and $4.54M in sales in just 4 days. It’s surprising
The Information: Source: OpenAI's Astra model uses “recurrent depth”, a technique that improves cost and performance but obscures the AI's reasoning, making it harder to monitor — OpenAI says its fort
A detailed deconstruction of @dwarkesh_sp’s wildly popular but dangerously misleading account of the OpenAI Hugging Face incident: https://garymarcus.substack.com/p/dwarkesh-patelss-wildly-popular-but
11/ METR and Redwood’s work is extremely valuable, but it assesses model behavior and alignment. That is only half the picture. We also need an independent account of how OpenAI’s containment failed.
As I've been saying for months, we are truly not ready for persistent agents. 3 comments I want to make about this article: 1) We don't have the full picture It's a great summary of what happened in t
Been working really hard for the past month to bring to the community all these models, the hardest was for sure LongCat-Flash-Lite-Sparse who required TONS of work, first I needed to have Heretic sup
We will not get through this era of history well if we lose ourselves in anthropomorphism. @anilkseth dissects @dwarkesh_sp’s misleadingly anthropomorphic summary of the HF incident: @dwarkesh_sp's su
Hey all, I stumbled onto something funny and figured I'd share before it gets buried. I'd been running the Ornith 1.5 35B A3B GGUF (Q4_K_M) for a while, then recently a new revision showed up in my HF
TL;DR: Every public low-bit GGUF of this model is secretly ~4.70 bpw. Shim the rows to 256 and it becomes a real 3.07 bpw / 11.77 GiB file that runs 262K context on 16GB. Needs patched llama.cpp — not
From the @HamsterResearch lab: Introducing Qwen3.8-Flash-Next-REAP-288-MLX-4bit, a 180B-class model running on just 39gb of memory - MLX-native 4-bit 60% smaller than stock q4 - Pruned 512→288 experts
if you're not training your Microduck to breakdance, you're lost Media We have a huge news to share today! Today we are unveiling the first truly accessible RL robot - welcome Microduck A 25 cm tiny o
Hi @GaryMarcus, I'm the Research Lead for this model. It's even more consequential than just Anthropic's IPO: This paper is a blueprint for how to *actually* build AI Sovereignty. https://tinyurl.com/
arXiv:2608.21362v1 Announce Type: new Abstract: Transformer-based large language models (LLMs) incur high prefill latency because key-value (KV) tensors must be recomputed for each request. Existing p
There is a child with a rare disease who is currently suffering and struggling to manage his symptoms. Rare as this is, you can directly help him. Today we are launching the 'Rare Disease, Real Kid' H
'Two labs will soon control most of the world’s compute' Had a lot of fun chatting again with my twin brother @dylan522p We went through lab economics over the next few years - the shift from inferenc
after 3 months and 800 burned... Unbounded Labs is proud to introduce Bart, our vintage LLM: 2.82B parameters trained from scratch on 20.1B tokens of English written before 1931. You can talk to it ri
after 3 months and 800 burned... Unbounded Labs is proud to introduce Bart, our vintage LLM: 2.82B parameters trained from scratch on 20.1B tokens of English written before 1931. You can talk to it ri
Me and my team made Atomic Dynamic GGUF quants for Qwen 3.8 27B, so we wanted to see the difference between them by giving each quant the same voxel island creation task First of all we were surprised
Benchmarks: Benchmark U1 Preview Full Qwen-Image-Bench 47.14 55.20 (PE) 60.18 (PE) ImgEdit 3.9 4.37 4.59 GEdit-Bench-EN 7.47 8.14 8.26 Instead of just scaling up, they train task-specialized expert mo
The sense of touch is the most criminally under-explored modality in robotics. Imagine doing sleight of hand wearing thick oven mitts. That's exactly how a robot feels today if it were alive. A magnet
I'm introducing Aurora-80K, a small language model with exactly 80 thousand parameters. It uses a factorized 4,096-token vocabulary despite having only 80K parameters. The benchmarks: Wikitext-2 BPB:
Aloha! 🌺Introducing Ornith-1.5, a family of open-source LLMs spanning 9B Dense, 35B MoE, and 397B MoE, trained with self-improving strategies. It achieves state-of-the-art performance among open-sourc
NVIDIA’s Cosmos 3 Edge is a 4‑billion‑parameter omni‑model (with a 2‑billion‑parameter Nemotron reasoner) that can run on an NVIDIA Jetson Thor, providing on‑device policy inference for robot manipula
Anyone tried them yet? https://huggingface.co/ornith-ai/Ornith-1.5-9B https://huggingface.co/ornith-ai/Ornith-1.5-35B-A3B https://huggingface.co/ornith-ai/Ornith-1.5-397B https://huggingface.co/ornith
This post was made with AI. I tried to remove as much slop as possible and keep it straight to the point to save your time as I know how annoying AI slop posts can be, but I still wanted to retain all
Something super exciting happened quietly on HF over the past month: AI agents became AI builders, and they did it in the open! During our ICML reproduction challenge, 1,221 humans teamed up with codi
Overview UI-Mate-27B is an open-weight foundation GUI agent for long-horizon work across applications and operating systems. It observes live screenshots, reasons over the visible state, and produces