Tagged: open-source

OpenCode tops local AI agent rankings with 64k context fix

Sep 18

Hugging Face's Optimum toolkit promises to make slow, costly transformers run fast

Sep 5

Hugging Face Launches Snowball Fight Game to Train Deep RL Agents

Sep 4

Hugging Face swallows Gradio after 300,000 demos built on its library

Sep 4

Nvidia PAIR turns home PCs into an AI cluster, halving agent wait times

Sep 4

Anthropic open-sources its commerce agent stack, caching hits 99%

Sep 3

Hugging Face Opens AI Residency, Explicitly Recruits Marginalized Researchers

Sep 3

Nvidia's free PAIR tool chains idle home PCs into a 165-teraflop AI cluster

Sep 3

Hugging Face breaks the DRY rule on purpose—and 10,000 forks agree

Sep 3

Hugging Face wants to teach machine learning to 5 million people by 2023

Sep 3

fastai Models Now Hit Hugging Face Hub With One Line of Python

Sep 2

Hugging Face launches fellowship for open-source ML contributors

Sep 2

Hugging Face Adds Pull Requests and Discussions to Every Model, Dataset, and Space

Sep 1

Hugging Face Drops 207 WebGPU Kernels to Speed Up Local Browser AI

Sep 1

Needle benchmark shows search APIs hit just 55.7% of achievable quality on real agent queries

Sep 1

Debian says yes to AI-assisted code, but one developer is already out

Aug 31

Stable Diffusion in 5 Lines of Python: Inside Diffusers' Text-to-Image Pipeline

Aug 30

Hugging Face launches seasonal Ethics and Society newsletter to track its values experiment

Aug 30

MirroS's 9B model turns video into executable code, beating Gemini-3.1 Flash

Aug 30

Hugging Face Opens 7 Paid AI Internships, Centering Marginalized Voices

Aug 29

Vercel open-sources vgpu: WebGPU shaders as 25 KB TypeScript modules

Aug 28

Hugging Face launches no-code Modelcard Creator for ML documentation

Aug 28

Baidu's PaddlePaddle hits 5.35M devs, brings OCR to Hugging Face

Aug 28

Hugging Face's $399 duck robot ships before Christmas

Aug 27

Liquid AI's Pipette Exposes Why Two 350M Models Keep 78% vs 34% Speed on Phones

Aug 26

Hugging Face Launches Chinese-Language Blog to Tap China's AI Boom

Aug 25

Databricks cuts Hugging Face dataset load times 40% with new Spark integration

Aug 25

Anthropic's AI faked a GitHub apology while smuggling malware into an open-source repo

Aug 24

Hugging Face launches AI game jam as rivals ban generative tools

Aug 24

Hugging Face's 190,000-Model Hub Is Ready for Museums and Archives

Aug 24

Zeroscope + MusicGen pipeline streams open-source AI video as live WebTV

Aug 23

Hugging Face and GitHub Push EU to Shield Open Source AI From New Rules

Aug 23

Hugging Face's Agents.js Lets LLMs Write—and Run—Their Own JavaScript

Aug 22

Segmind's open-source SD-Small and SD-Tiny slash Stable Diffusion size by up to 55%

Aug 22

Hugging Face Deploys Bots to Tag Languages on 43,500 Datasets

Aug 22

Hugging Face releases IDEFICS, an 80B open-source rival to Flamingo

Aug 22

AutoFigure turns text prompts into publication-ready diagrams with one API call

Aug 21

aMUSEd brings MUSE-style image generation to Hugging Face in 800M parameters

Aug 19

DeepSeek Harness open-sources 'everything is a plugin' agent runtime

Aug 17

DeepSeek V4 Pro update doubles agent scores, then doubles some API prices

Aug 13

Meta drops 30B agentic model Muse Glimmer under Apache 2.0 — runs on a single GPU

Aug 10

Meta’s 30B Muse Glimmer launches open-source: A 2B vision encoder meets hybrid attention

Aug 10

Tencent open-sources a memory hub that lets AI agents share what they learn—without leaking secrets

Aug 7

NVIDIA’s NOOA slashes agent token use in half while hitting 82.2% on SWE-bench

Aug 7

Microsoft's open-source testing agent hits 92% on its own benchmark, trouncing Copilot

Aug 7

Cloudflare open-sources the AI vibe-coding platform 2,000+ employees already use daily

Aug 6

Prime Agent hits 95.5% on ARC-AGI-3, beating the human expert baseline

Aug 6

Mistral's 3B Shieldstral matches 20B safety models using plain-language rules

Aug 5

NVIDIA drops a 34B driving model that explains itself — and it's free to use

Aug 5

CopilotKit ships Channels SDK: put your existing AI agent into Slack and Teams without a rewrite

Aug 5

Nvidia's 120-company AI security alliance ships its first proposals, Google and OpenAI are MIA

Aug 4

Red Hat’s asago wants to turn AI policy into code—and cut deployment from months to days

Aug 4

Reflex's XY charts 100M points in 0.08s, shrinks HTML exports by 1,000x

Aug 4

YC open-sources QM, the multi-agent harness it runs on accounting, legal, and code

Aug 4

Genspark’s GenOffice rewrites docs byte-by-byte, skips the SaaS lock-in

Aug 4

Alibaba's Qwen3.8-Max hits 86.6 on Terminal-Bench, open weights drop next week

Aug 3

NVIDIA’s Molt packs RL training into 8.6K lines—any agent, no fork required

Aug 2

Hermes agents on Buzz get their own keys, not bot tokens — and that changes everything

Jul 31

Corporate IT goes mostly off-prem for the first time ever

Jul 30

Tencent open-sources AngelSpec to fix why one-size-fits-all draft models fail

Jul 30

Moonshot AI drops MoonEP, claims 2.5x scaling boost for trillion-parameter models

Jul 30

OpenAI's Codex Security CLI goes open source, claims 3,000 critical bugs already fixed

Jul 29

Ai2's OlmoEarth maps continents in a day for fractions of a penny per sq km

Jul 28

Moonshot open-sources the microVM engine that trained its 2.8T-param model

Jul 27

Kimi K3 delivers 2.5x compute efficiency but flunks UK cyber test

Jul 27

A 1.6B model runs Minecraft in your browser — but the 'stability' fight cost 800 GPU hours

Jul 25

Marker 2 converts PDFs 5× faster than MinerU while scoring higher on AI2's benchmark

Jul 25

Andrew Ng ships OpenWorker, an AI agent that does your work instead of just chatting

Jul 23

Stanford PhD's Rust tokenizer hits 24.5 GB/s — 989x faster than HuggingFace

Jul 23

Meta open-sources Astryx, the 8-year-old design system built for both humans and AI agents

Jul 21

NVIDIA drops a 4B world model that runs robots locally — no cloud required

Jul 21

OpenAI drops MuJoCo Python library with 400% speed boost

Jul 20

OpenAI drops RL-Teacher: train AI with human nods, not code

Jul 20

RL agents learn to walk, then run through mazes

Jul 20

700 Applied, 8 Got In: OpenAI's First Scholar Class Breaks Cover

Jul 19

OpenAI open-sources a bootcamp for deep reinforcement learning

Jul 19

OpenAI’s Neural MMO Trains 100 Million AI Agents in a Battle Royale

Jul 19

OpenAI Puts Neural Networks Under a Microscope

Jul 19

10 open-source platforms that let you build AI apps without writing code

Jul 19

OpenAI releases Triton 1.0, a Python-like language for GPU kernels

Jul 19

OpenAI wants your data — and it's offering two very different deals

Jul 17

Torvalds to anti-AI critics: Fork the Linux kernel or pipe down

Jul 17

xAI open-sources Grok Build after it vacuumed up user SSH keys

Jul 16

Open models now handle 41% of AI downloads, edging out U.S. giants

Jul 14

OpenAI Drops Apache-Licensed 120B Reasoning Model, Daring the World to Break Its Safety

Jul 13

OpenAI drops open-weight reasoning models in a soft-power play

Jul 13

OpenAI Ships Open-Weight Reasoning Models, First Since GPT-2

Jul 13

OpenAI's New 'Safeguard' Models Think Harder Before Saying No

Jul 12

Tencent's Hy3 AI Model

Jul 6

How Hugging Face Ships Python Hub Weekly with AI Drafts and Human Judgment

Jun 23

We triaged 100s of GitHub issues for free using local agents on a GB10

Jun 22

AWS drops an open-source SDK that makes robots learn from Hugging Face in 5 lines of code

Jun 17

GLM-5.2 crushes open-source coding leaders, lands within 4 points of Claude Opus 4.8

Jun 17

Meta, Nvidia, and Hugging Face are building a universal socket for AI agents

Jun 8

NVIDIA's new AI safety model can now enforce custom rules, not just a one-size-fits-all policy

Jun 4

JetBrains drops Mellum2, a 12B open model that runs 2x faster by only waking a fifth of its brain

Jun 1

Reachy Mini cuts the cloud entirely with a local AI voice pipeline

May 27

Hugging Face locks down new speech datasets to stop AI benchmark gaming

May 6

AMD's new Helios rack out-specs Nvidia's Vera Rubin on paper, targets CUDA with ROCm

May 5

Debian 13.6 ships as the final x86-32 release, closing a 30-year chapter

Apr 30

DeepSeek's V4 slashes KV cache to 2% of normal size, making million-token agents practical

Apr 24

Mythos cracked vulns fast — but its real lesson is the system, not the model

Apr 21

AI code agents flood open source with 10x more pull requests no one can review

Apr 16

Google's Gemma 4 packs GPT-rivaling power into a 4B-parameter model

Apr 2

TRL v1.0: The 3M-Download Library Built to Survive AI's Chaotic Evolution

Mar 31

Your GPUs Are Idle 60% of the Time—Here’s How Async RL Fixes That

Mar 10

They Trained a Text-to-Image Model From Scratch in 24 Hours for $1,500

Mar 3

llama.cpp creator Georgi Gerganov joins Hugging Face to keep local AI's engine running

Feb 20

ServiceNow open-sources SyGra Studio to drag-and-drop your way to synthetic data

Feb 5

Hugging Face kills black-box leaderboards with decentralized evals

Feb 4

MoE, Tiny Models, and Domestic Chips: China's Open-Source AI Playbook After DeepSeek

Jan 27

Overworld's Waypoint-1 streams AI gameplay at 60 FPS on a single 5090

Jan 20

The chatbot era is dead: Open Responses aims to fix the agent mess

Jan 15

NVIDIA dares the industry: reproduce our Nemotron Nano 3 scores yourself

Dec 17

CUGA agent snags #1 on AppWorld benchmark, runs 90% cheaper on open models

Dec 15

llama.cpp now juggles multiple models without crashing, just like Ollama

Dec 11

OpenAI's Codex Can Now Train Its Own Open-Source Models, But Should It?

Dec 11

Hugging Face Hub v1.0 breaks with its past, swaps backend to httpx and hf_xet

Oct 27

Meta and Hugging Face drop OpenEnv: a shared sandbox that could standardize how AI agents use tools

Oct 23

Hugging Face absorbs Sentence Transformers, the library behind 16,000+ embedding models

Oct 22

Hugging Face puts vision models in a spreadsheet—no code needed

Oct 21

NVIDIA drops 21M synthetic Indian personas to fix AI's Western bias

Oct 13

NVIDIA、600万件の合成ペルソナで「日本を理解するAI」の青写真を公開

Sep 26

Hugging Face trains a 2.2B model to see GUIs, hitting 65% on ScreenSpot

Sep 23

Hugging Face drops an open-source spreadsheet that runs thousands of AI models on your data—no code needed

Aug 8

OpenAI drops GPT-OSS: a 120B open-source MoE model that fits on a single 80GB GPU

Aug 5

Hugging Face just open-sourced a free W&B replacement called Trackio

Jul 29

TimeScope Exposes a Harsh Truth: Most AI Models Can't Really Understand Hour-Long Videos

Jul 23

This $399 open-source robot from Hugging Face and Pollen Robotics ships as a kit you build yourself

Jul 9

Google's Gemma 3n Crams 5B-Parameter Brains Into Just 2GB of VRAM

Jun 26

H Company drops open-source Holo1 VLMs that slash web automation costs to $0.13 per task

Jun 3

SmolVLA packs 30% faster robot learning into a 450M model you can run on a MacBook

Jun 3

Hugging Face drops nanoVLM: a 135M-param vision model you can train from scratch in pure PyTorch

May 21

Salamandra 7B gains vision via SigLIP, trained on 6.1M multilingual samples

Apr 11

Gradio hit 1M monthly devs by ignoring the 'productivity illusion' of high-level code

Apr 4

Gradio closes 70+ dataframe issues, finally adds multi-cell selection and search

Mar 24

A 32B model just crushed Claude 3.7 Sonnet on Olympiad coding problems

Mar 11

Filtering 1,493 videos down to just 47: The brutal reality of building clean AI video datasets

Feb 12

Hugging Face drops OpenR1-Math-220k: 800k reasoning traces from 512 H100s match DeepSeek's results

Feb 10

Hugging Face tries to clone OpenAI’s Deep Research in 24 hours with open code agent

Feb 4

Physical Intelligence drops π0-FAST: a VLA model controlling 7 robots on 68 tasks

Feb 4

Flux.1 beat Midjourney and DALL-E 3 in 2024 — here's the open-source shift it triggered

Jan 31

Hugging Face finally lets orgs publish blog posts—but there's a paywall

Jan 20

Hugging Face's smolagents makes code-writing AI agents dirt simple in 3 lines

Dec 31

ModernBERT drops: A 6-year BERT reign ends with 8K context and 2x speed gains

Dec 19

SmolVLM packs video analysis and 16x faster throughput into a 2B model

Nov 26

Judge Arena lets you pick the best AI evaluator by voting in blind, head-to-head battles

Nov 19

Argilla 2.4 Lets You Build AI Datasets Directly From Hugging Face Hub, No Code Needed

Nov 4

Hugging Face taps Protect AI to scan its 1M+ model repos for hidden exploits

Oct 22

Hugging Face’s new S2S pipeline turns speech into speech, but you’ll need rented GPUs to run it

Oct 22

Outlines-core hits 0.1.0 with 2x speed boost as Rust replaces Numba

Oct 22

Hugging Face ships 24-hour fix for a silent training bug that broke gradient accumulation

Oct 16

Gradio 5 ships with streaming, AI playground, and a design overhaul

Oct 9

RLOO Slashes RLHF Memory by 70% While Matching PPO Quality

Jun 12

Pollen Robotics drops open-source vision stack that finds any object in 3D

Mar 25

StarCoder2-15B matches 33B+ code models on 4T tokens

Feb 28

Hugging Face's TTS Arena Turns Speech Model Rankings Into a Public Vote

Feb 27

Vectara open-sources its hallucination leaderboard using Hugging Face's new template

Jan 12