This GPT-5 system card explains how a unified model routing system powers fast and smart responses using gpt-5-main, gpt-5-thinking, and lightweight versions like gpt-5-thinking-nano, optimized for different tasks and developer use.
The state of the research Diff Risk Score (DRS) is an AI-powered technology built at Meta that predicts the likelihood of a code change causing a production incident, also known as a SEV. Built on a fine-tuned Llama LLM, DRS evaluates code changes and metadata to produce a risk score and highlight potentially risky code [...] Read More... The post Diff Risk Score: AI-driven risk-aware software dev
Today, OpenAI for Government is announcing a new partnership with the U.S. General Services Administration (GSA) to launch a transformative initiative. For the next year, ChatGPT Enterprise will be available to the entire federal executive branch workforce at essentially no cost.
AI’s next frontier isn’t just about capability—it’s about who gets to use it. Our mission to put AI in the hands of as many people as possible is what drives us. Today’s release of our most capable open-weights models is a major step forward that makes advanced AI more open, flexible, and accessible worldwide.
Official Artificial Analysis technical update and publication covering Articles.
We introduce gpt-oss-120b and gpt-oss-20b, two open-weight reasoning models available under the Apache 2.0 license and our gpt-oss usage policy.
How modern DevOps teams are replacing legacy CI/CD pipelines and accelerating transformation using AI automation.
Official technical announcement and publication from Hugging Face covering Welcome GPT OSS, the new open-source model family from OpenAI!.
We’re releasing gpt-oss-120b and gpt-oss-20b—two state-of-the-art open-weight language models that deliver strong real-world performance at low cost. Available under the flexible Apache 2.0 license, these models outperform similarly sized open models on reasoning tasks, demonstrate strong tool use capabilities, and are optimized for efficient deployment on consumer hardware.
In this paper, we study the worst-case frontier risks of releasing gpt-oss. We introduce malicious fine-tuning (MFT), where we attempt to elicit maximum capabilities by fine-tuning gpt-oss to be as capable as possible in two domains: biology and cybersecurity.
Access OpenAI’s gpt-oss-120B on Together AI: Apache-2.0 open-weight model with serverless & dedicated endpoints, $0.50/1M in, $1.50/1M out, 99.9% SLA.
Official Decagon technical update and publication covering From WISMO to upsells: How retail brands are using AI agents.
Official technical announcement and publication from Hugging Face covering Measuring Open-Source Llama Nemotron Models on DeepResearch Bench.
Zainab Fatmi and Professors Stefan Kiefer and David Parker have developed a new method for simplifying probabilistic models in formal verification
We build ChatGPT to help you thrive in all the ways you want. Learn how we’re improving support for tough moments, have rolled out reminders to take breaks, and are working on better life advice, all guided by expert input.
Official technical announcement and publication from Hugging Face covering 📚 3LM: A Benchmark for Arabic LLMs in STEM and Code.
Discover how Figma is transforming digital design with AI. David Kossnick shares how tools like Figma Make empower teams to prototype, collaborate, and build with AI—reshaping workflows for designers, developers, and non-technical creators alike.
Official Mistral AI technical update and publication covering Unlocking the potential of vision language models on satellite imagery through fine-tuning.
We’re launching Stargate Norway—OpenAI’s first AI data center initiative in Europe under our OpenAI for Countries program. Stargate is OpenAI’s overarching infrastructure platform and is a critical part of our long-term vision to deliver the benefits of AI to everyone.
Official technical announcement and publication from Hugging Face covering Implementing MCP Servers in Python: An AI Shopping Assistant with Gradio.
Official LMSYS Chatbot Arena release and benchmark update covering GLM-4.5 Meets SGLang: Reasoning, Coding, and Agentic Abilities.
Discover how Intercom built a scalable AI platform with 3 key lessons—from evaluations to architecture—to lead the future of customer support.
Official Mistral AI technical update and publication covering Announcing Codestral 25.08 and the Complete Mistral Coding Stack for Enterprise.
Introducing study mode in ChatGPT, a new learning experience that helps you work through problems step by step, guiding students with questions, scaffolding, and feedback for deeper learning.
Official technical announcement and publication from Hugging Face covering Introducing Trackio: A Lightweight Experiment Tracking Library from Hugging Face.
Official technical announcement and publication from Together AI covering VirtueGuard: Enterprise-Grade AI Security and Safety Now on Together AI.
Together Evaluations is a flexible framework for benchmarking LLMs using strong open-source models as judges. Skip manual labeling and rigid metrics—get fast, customizable insights into model quality for your specific tasks.
Official technical announcement and publication from Hugging Face covering Parquet Content-Defined Chunking.
Official technical announcement and publication from Hugging Face covering Say hello to `hf`: a faster, friendlier Hugging Face CLI ✨.
Unlock agentic coding with Qwen3-Coder on Together AI: 256K context, SWE-bench rivaling Claude Sonnet 4, zero-setup instant deployment.
Official LMSYS Chatbot Arena release and benchmark update covering SpecForge: Accelerating Speculative Decoding Training for SGLang.
We are delighted to announce the appointment of Graham Cormode as the Professor of Computer Science. He will take up the post on 1 January 2026 and will hold a Professorial Fellowship at St John's College.
Discover how Outtake uses GPT-4.1 and OpenAI o3 to power AI agents that detect and resolve digital threats 100x faster than before.
Official Groq (LPUs) technical update and publication covering Groq Applauds Trump’s AI Action Plan, Accelerates Global AI Stack.
We’re hosting our third annual OpenAI DevDay on October 6, 2025 at Fort Mason in San Francisco.
Official technical announcement and publication from Hugging Face covering TimeScope: How Long Can Your Video Large Multimodal Model Go?.
As part of our Executive Function series, Model ML CEO Chaz Englander discusses how AI-native infrastructure and autonomous agents are transforming financial services workflows.
Official technical announcement and publication from Hugging Face covering Fast LoRA inference for Flux with Diffusers and PEFT.
Vector’s latest annual report showcases research advancements and industry partnerships that strengthen Canada’s AI leadership Vector bridges AI research and application, translating cutting-edge science into solutions that benefit Canadians. Over […] The post Vector Institute 2024-25 annual report: Where AI research meets real-world impact appeared first on Vector Institute for Artificial Intelli
OpenAI and Penda Health debut an AI clinical copilot that cuts diagnostic errors by 16% in real-world use—offering a new path for safe, effective AI in healthcare.
Analysis provides insights into ChatGPT’s impact on the economy. OpenAI also launches new research collaboration to study AI’s broader effects on the labor market and productivity.
You can now connect your favorite MCP servers to Devin.
Oracle and OpenAI have entered an agreement to develop 4.5 gigawatts of additional Stargate data center capacity in the U.S. This investment will create new jobs, accelerate America’s reindustrialization, and help advance U.S. AI leadership. It also marks a major milestone for Stargate, OpenAI’s AI infrastructure platform and long-term vision to deliver the benefits of AI to everyone.
Official Mistral AI technical update and publication covering Our contribution to a global environmental standard for AI.
Official technical announcement and publication from Hugging Face covering Accelerate a World of LLMs on Hugging Face with NVIDIA NIM.
OpenAI partners with the UK Government to boost AI adoption, drive economic growth, and enhance public services for a thriving AI ecosystem in the UK.
I’ve always considered myself a pragmatic technologist—someone who loves technology not for its own sake, but for the direct impact it can have on people’s lives. That’s what makes this job so exciting, since I believe AI will unlock more opportunities for more people than any other technology in history. If we get this right, AI can give everyone more power than ever.
Official LMSYS Chatbot Arena release and benchmark update covering Deploying Kimi K2 with PD Disaggregation and Large-Scale Expert Parallelism on 128 H200 GPUs.
OpenAI is launching an initial $50 million fund that supports nonprofit and community organizations, informed by the independent OpenAI Nonprofit Commission report.
Official technical announcement and publication from Hugging Face covering Arc Virtual Cell Challenge: A Primer.
Introducing ChatGPT agent: it thinks and acts, using tools to complete tasks like research, bookings, and slideshows—all with your guidance.
ChatGPT agent System Card: OpenAI’s agentic model unites research, browser automation, and code tools with safeguards under the Preparedness Framework.
Official technical announcement and publication from Hugging Face covering Five Big Improvements to Gradio MCP Servers.
At OpenAI, we build tools to help people solve hard problems—including nonprofits working on the frontlines of their communities. The OpenAI Academy is teaming up with the Walton Family Foundation, Emerson Collective, and a network of local nonprofit organizations to host the Nonprofit Jam—a one-day, nationwide event bringing together more than 1,000 nonprofit leaders across 10 locations.
Invideo AI uses OpenAI’s GPT-4.1, gpt-image-1, and text-to-speech models to transform creative ideas into professional videos in minutes.
Official technical announcement and publication from Hugging Face covering Consilium: When Multiple LLMs Collaborate.
The Board of Directors thanks the members of the independent OpenAI Nonprofit Commission for their extensive work and engagement.
Official technical announcement and publication from Hugging Face covering Back to The Future: Evaluating AI Agents on Predicting Future Events.
Official LMSYS Chatbot Arena release and benchmark update covering Accelerating SGLang with Multiple Token Prediction.
Official Mistral AI technical update and publication covering Le Chat dives deep..
The Erlangen AI Hub has rapidly become a major force in UK artificial intelligence research
Edd Salkield's research identified and analysed a critical flaw in the latest generation of satellite communication standards
Official technical announcement and publication from Hugging Face covering Ettin Suite: SoTA Paired Encoders and Decoders.
Official LMSYS Chatbot Arena release and benchmark update covering How to support new VLMs into SGLang: A Case Study with NVILA.
Official technical announcement and publication from Hugging Face covering Migrating the Hub from Git LFS to Xet.
Choosing a speech API isn’t just about specs. It’s about what happens when real-life audio hits the system.
Choosing a speech API isn’t just about specs. It’s about what happens when real-life audio hits the system.
ChatGPT is designed to be useful, trustworthy, and adaptable—so you can make it your own.
Cognition has signed a definitive agreement to acquire Windsurf, the agentic IDE.
Official LMSYS Chatbot Arena release and benchmark update covering Cost Effective Deployment of DeepSeek R1 with Intel® Xeon® 6 CPU on SGLang.
OpenAI joins the EU Code of Practice, advancing responsible AI while partnering with European governments to drive innovation, infrastructure, and economic growth.
Hierarchical modeling
Leading researchers from Vector are presenting cutting-edge work at this year’s International Conference on Machine Learning (ICML), taking place July 13-19, 2025 in Vancouver, Canada and through virtual platforms. With […] The post Vector researchers tackle real-world AI challenges at ICML 2025 appeared first on Vector Institute for Artificial Intelligence .
Official technical announcement and publication from Hugging Face covering Kimina-Prover: Applying Test-time RL Search on Large Formal Reasoning Models.
Official technical announcement and publication from Hugging Face covering Asynchronous Robot Inference: Decoupling Action Prediction and Execution.
Official technical announcement and publication from Hugging Face covering ScreenEnv: Deploy your full stack Desktop Agent.
Official technical announcement and publication from Hugging Face covering Building the Hugging Face MCP Server.
Official Mistral AI technical update and publication covering Upgrading agentic coding capabilities with the new Devstral models.
Mihaela Stoian, a final-year DPhil student in Computer Science, has been awarded the Oxford PhD Runner-up Prize by G-Research.
Official Groq (LPUs) technical update and publication covering Groq Named Top 10 National Security Tech Company in U.S..
Official technical announcement and publication from Hugging Face covering Reachy Mini - The Open-Source Robot for Today's and Tomorrow's AI Builders.
Official technical announcement and publication from Hugging Face covering Upskill your LLMs With Gradio MCP Servers.
Official technical announcement and publication from Hugging Face covering Creating custom kernels for the AMD MI300.
Official LMSYS Chatbot Arena release and benchmark update covering slime: An SGLang-Native Post-Training Framework for RL Scaling.
OpenAI partners with the American Federation of Teachers to launch a 5-year initiative equipping 400,000 K-12 educators to lead AI innovation in classrooms.
Official technical announcement and publication from Hugging Face covering Efficient MultiModal Data Pipeline.
Official technical announcement and publication from Hugging Face covering Three Mighty Alerts Supporting Hugging Face’s Production Infrastructure.
Official technical announcement and publication from Hugging Face covering SmolLM3: smol, multilingual, long-context reasoner.
Official LMSYS Chatbot Arena release and benchmark update covering OME: Revolutionizing LLM Infrastructure with Model-Driven Architecture.
Official Groq (LPUs) technical update and publication covering Groq Launches European Data Center Footprint in Helsinki, Finland.
Official technical announcement and publication from Hugging Face covering Announcing NeurIPS 2025 E2LM Competition: Early Training Evaluation of Language Models.
Official Mistral AI technical update and publication covering Announcing AI for Citizens.
Learn how Genspark built a $36M ARR AI product in 45 days—with no-code agents powered by GPT-4.1 and OpenAI Realtime API.
Official technical announcement and publication from Hugging Face covering Training and Finetuning Sparse Embedding Models with Sentence Transformers.
VGG team wins top prize at CVPR 2025 with 3D vision breakthrough
Today, OpenAI, in partnership with Mandala Partners, is sharing the OpenAI AI Economic Blueprint for Australia. At a time when boosting productivity has emerged as a national priority for Australia, the Blueprint provides a clear, actionable plan for how Australia can unlock the full economic and social potential of artificial intelligence.
Official technical announcement and publication from Hugging Face covering Welcome the NVIDIA Llama Nemotron Nano VLM to Hugging Face Hub.
The year is 2025. Coding agents aren't magic, but they're about the closest thing we have. We've noticed some engineers, in particular at the senior-to-staff level, finding success faster than others. Here we share some top lessons sourced from the experience of our customers and ourselves.
Retell AI is transforming the call center with AI voice automation powered by GPT-4o and GPT-4.1. Its no-code platform enables businesses to launch natural, real-time voice agents that cut call costs, boost CSAT, and automate customer conversations—without scripts or hold times.
Official technical announcement and publication from Hugging Face covering Gemma 3n fully available in the open-source ecosystem!.
6584 articles sourced historically · 100 per page