Latest open artifacts (#19): Qwen 3.5, GLM 5, MiniMax 2.5 — Chinese labs' latest push of the frontier
Qwen 3.5, GLM 5, and MiniMax 2.5: Chinese labs release latest frontier models. Interconnects covers recent advances from major Chinese AI players.
Qwen 3.5, GLM 5, and MiniMax 2.5: Chinese labs release latest frontier models. Interconnects covers recent advances from major Chinese AI players.
OpenAI releases System Card for GPT-5.3 Instant, documenting model capabilities, limitations, and safety measures. Official technical documentation with no release date or performance metrics disclosed.
The France Digitale/EY Barometer 2024 demonstrates the resilience of France's startup ecosystem despite economic and geopolitical instability. The annual study documents continued sector growth amid macroeconomic challenges.
OpenAI announces mental health safety updates: parental controls, trusted contacts feature, improved distress detection, and litigation developments. No technical details or impact metrics disclosed in excerpt.
Google DeepMind releases Nano Banana 2, an image generation model combining advanced world knowledge, subject consistency, and Flash-speed performance. Production-ready.
Nathan Lambert examines the actual importance of distillation for Chinese LLMs, responding to Anthropic's post on 'distillation attacks.' The article investigates whether model distillation remains a critical technique for Chinese developers facing restrictions on closed-model access.
Seedream 5.0 introduces multi-step reasoning, example-based editing, and deep domain knowledge to image generation. Guide to key capabilities.
Gradio introduces gr.HTML, a component enabling custom HTML content integration into web applications. This feature simplifies building complex user interfaces while staying within the Gradio ecosystem.
Nathan Lambert analyzes the persistent gap between open-source and proprietary models. He examines distillation, innovation cycles, winning strategies for open models, specialization, and structural gaps in the industry.
Google DeepMind launches National Partnerships for AI initiative in India to accelerate scientific discovery and education through AI. The program aims to tailor AI tools to local challenges and strengthen research capabilities.
Hugging Face introduces custom kernels for all through Codex and Claude. The initiative aims to democratize access to compute optimizations for AI models by using AI-assisted code generation to create specialized kernels tailored to different hardware.
AI will not automatically reduce legal services costs. The article applies the 'AI as Normal Technology' framework to the legal sector, questioning the assumption that AI automation will systematically drive down prices.
Google DeepMind announces Gemini 3 Deep Think, an update to its specialized reasoning mode designed to tackle modern science, research and engineering challenges.
Hugging Face introduces SyGra Studio, a platform for creating and managing knowledge graphs. The tool enables building, visualizing, and deploying complex data structures to enhance AI model reasoning capabilities.
Hugging Face analyzes the evolution of the global open-source AI ecosystem, particularly DeepSeek's impact and future trends. The article examines how open-source models are transforming AI accessibility and competitiveness.
Critique of Moravec's paradox relevance, the famous claim that tasks easy for humans are hard for AI and vice versa. The article questions the validity and usefulness of this principle in current context.
OpenAI releases EU Economic Blueprint 2.0 with new data, partnerships, and initiatives to accelerate AI adoption, skills development, and economic growth across Europe.
Hugging Face used Claude to generate CUDA kernels and train open-source models. The experiment demonstrates Claude's capabilities in low-level code and GPU optimization.
Analysis of architectural choices in China's open-source AI ecosystem beyond DeepSeek. Examination of models, infrastructures and development strategies of local players competing internationally.
Hugging Face shares practical insights on agentic reinforcement learning training for GPT open-source models. Retrospective on challenges and solutions for scaling RL deployment across free models.
OpenAI releases data-driven report on ChatGPT adoption across industries: usage trends, top tasks, departmental patterns, and future of AI at work. No specific metrics disclosed in excerpt.
Higgsfield leverages OpenAI's GPT-4.1, GPT-5, and Sora 2 to convert simple ideas into cinematic, social-first videos. The platform automates video content generation from minimal inputs for creators.
OpenAI launches Edu for Countries, a program for governments to integrate AI into national education systems and develop future-ready workforces. Initiative announced without technical details or impact metrics.
OpenAI releases report showing stark differences in advanced AI adoption across countries and announces new initiatives to help nations capture productivity gains. No specific data or program details provided in excerpt.
Hugging Face introduces Differential Transformer V2, an architecture improving long-range dependency handling through differential attention mechanisms. Version 2 optimizes performance and training stability compared to V1.
Overworld introduces Waypoint-1, a real-time interactive video diffusion model. The system generates high-resolution videos with low latency, enabling direct user interaction during generation.
OpenAI outlines a multi-channel business model (subscriptions, API, ads, commerce, compute) that scales with AI advancement. Growth hinges on deepening ChatGPT adoption and monetizing intelligence across multiple revenue streams.
Hugging Face publishes a guide on open responses, explaining principles and practices for increased transparency in the AI ecosystem. The article covers best practices for communication and documentation in open-source projects.
OpenAI launches RFP to strengthen U.S. AI supply chain through domestic manufacturing, job creation, and infrastructure scaling. No technical specifications or budget details disclosed in announcement.
Zenken rolls out ChatGPT Enterprise company-wide to boost sales performance, cut preparation time, and increase proposal success rates. AI-powered workflows enable a lean team to deliver more personalized and effective customer engagement.
OpenAI launches Healthcare solution with HIPAA compliance for enterprises, aiming to reduce administrative burden and support clinical workflows.
Tolan built a voice-first AI companion using GPT-5.1, combining low-latency responses, real-time context reconstruction, and memory-driven personalities for natural conversations.
Hugging Face introduces Falcon-H1-Arabic, a hybrid architecture model for Arabic combining transformers and alternative approaches. The model advances Arabic language AI capabilities with an innovative architectural design.
Hugging Face introduces AprielGuard, a guardrail system to enhance safety and adversarial robustness in modern LLM systems. The tool detects and mitigates adversarial attacks and harmful outputs in real-time.
OpenAI reaches one million customers globally. Post highlights adoption by PayPal, Virgin Atlantic, BBVA, Cisco, Moderna, and Canva, but lacks technical details or measurable impact metrics.
OpenAI releases AI literacy guides for teens and parents on responsible ChatGPT use, critical thinking, and healthy boundaries. Expert-vetted resources for navigating AI safely and thoughtfully.
OpenAI and U.S. Department of Energy sign memorandum of understanding to deepen collaboration on AI and advanced computing for scientific discovery. Agreement builds on existing work with national laboratories and establishes framework for applying AI to high-impact research across DOE ecosystem.
OpenAI releases an addendum to GPT-5.2 System Card for GPT-5.2-Codex, a code-generation variant. The document outlines capabilities, limitations, and safety measures. No performance metrics or availability date disclosed in excerpt.
OpenAI releases GPT-5.2-Codex, its most advanced coding model featuring long-horizon reasoning, large-scale code transformations, and enhanced cybersecurity capabilities. No technical details, benchmarks, or availability date disclosed in the announcement.
Google DeepMind announces Gemini 3 Flash, a model delivering frontier intelligence with increased speed and reduced costs compared to previous versions.