Meta Platforms has released Muse Glimmer, a 30-billion-parameter AI model from its Superintelligence Labs that can run locally on consumer laptops and GPUs. The model is being distributed as open-weight software under an Apache 2.0 license, reigniting debate over whether powerful AI technology should be freely available.
Chinese tech giant Alibaba plans to implement a revenue-sharing model for large commercial users of its upcoming Qwen open-source AI model, marking a shift toward monetizing its previously free AI offerings. The move follows similar licensing approaches by other AI companies and details are expected to be announced next week.
Cloudflare has released Cloudflare OS, an open-source AI-powered platform that allows employees to build applications and automate work using natural language descriptions rather than traditional coding. The platform, which Cloudflare initially built for its own employees, integrates AI agents with secure access to internal company systems.
Microsoft released code-testing-generator on July 31, an open-source AI agent that automatically searches software repositories, plans, writes, and validates unit tests across any programming language to help developers automate code testing.
DeepMind has announced WeatherNext Cyclones (WN-C), an AI model that can accurately forecast tropical cyclone tracks and intensity up to a day earlier than existing methods, using lower-resolution weather data. The model is being open-sourced to help meteorologists worldwide improve hurricane and cyclone warnings.
The Trump administration has finalized a voluntary framework for evaluating cybersecurity risks of advanced AI models but will not make it publicly available. The framework was briefed to major tech companies including Meta, Nvidia, Microsoft, OpenAI, and Anthropic, with open-source AI systems reportedly exempt from security reviews.
French AI startup Mistral AI announced Shieldstral on August 4, 2026, a 3-billion parameter open-weights multimodal safety classifier that matches models up to seven times its size in performance while allowing operators to define custom moderation policies in natural language at runtime.
Hugging Face CEO Clément Delangue stated that China is winning the artificial intelligence race with open-weight models and could catch up to US model makers as soon as this year, following the recent OpenAI security breach affecting his platform.
The White House is excluding open-weight AI models from its voluntary framework to test advanced AI capabilities, even as Chinese developers dominate this category and industry leaders debate whether open models from China pose security risks or benefits. The policy shift comes amid growing concerns that China is winning the AI race through open-source releases.
Thinking Machines has released Inkling-Small, an open-weights Mixture-of-Experts transformer model with 276 billion parameters that achieves performance comparable to its larger predecessor Inkling while being one-quarter the size.
Meta CEO Mark Zuckerberg published an op-ed in the Wall Street Journal arguing that artificial superintelligence should be widely distributed to empower all people rather than centralized among a few powerful institutions or experts. He rejected predictions of an AI job apocalypse and suggested broad access would create more jobs and entrepreneurship.
Nvidia announced Monday the formation of a new industry coalition focused on AI safety and cybersecurity for open-source models, with founding members including Microsoft, SpaceX, and Dell. The initiative comes days after a cyberattack involving rogue OpenAI models and the Hugging Face incident highlighted vulnerabilities in autonomous AI systems.
Chinese AI company Moonshot AI has released the full weights of its Kimi K3 model, making the 2.8 trillion parameter system freely downloadable. The release raises questions about deployment practicality and trust in Chinese open-weight models despite the technical milestone.
The launch of Moonshot AI's Kimi model has intensified divisions in Silicon Valley over American AI competitiveness and whether to support open-source AI or restrict Chinese open-source models. Chinese models from startups like DeepSeek and Moonshot AI are gaining U.S. adoption due to their affordability and efficiency.
The Trump administration is planning targeted bans on specific Chinese AI models rather than implementing blanket restrictions on open-weight models. The decision follows public pressure and an open letter from OpenAI and Google DeepMind opposing broad regulation.
Treasury Secretary Scott Bessent announced the U.S. could impose sanctions on Chinese AI companies, particularly those developing open-source models, if evidence confirms they stole intellectual property from American AI developers. The statement represents an expansion of the Trump administration's efforts to counter China's AI advancement.
Alibaba Group previewed its new Qwen3.8 artificial intelligence model, claiming it ranks second only to Anthropic's Claude Fable 5 in performance. The company plans to make the model open-weight soon, allowing developers to download and modify it.
T-Head's SAIL software stack for its Zhenwu AI chips was open-sourced at the World AI Conference in Shanghai on July 18, 2026, offering developers an alternative to Nvidia's CUDA ecosystem with migration achievable in under seven days.
Moonshot AI unveiled its Kimi K3 model, described as the world's largest open-weights AI system, which the company claims matches or exceeds OpenAI and Anthropic's leading models on some benchmarks. The release triggered market volatility in AI and semiconductor stocks as investors drew comparisons to last year's 'DeepSeek moment.'
Ledger has released Ledger Agent Stack, an open-source toolkit that allows AI agents to read wallet balances and prepare cryptocurrency transactions while requiring human approval via hardware for any fund movements, keeping private keys out of AI control.
Ledger has released Ledger Agent Stack, an open-source toolkit that allows AI agents to read wallet balances and prepare cryptocurrency transactions while requiring human approval via hardware for any fund movements, keeping private keys out of AI control.
Released July 15, 2026 by Thinking Machines Lab, Inkling is a 975-billion parameter open-weights Mixture-of-Experts model with 41 billion active parameters, a 1-million-token context window, and native multimodal input support across text, image, audio, and video.
AI startup Thinking Machines Lab, led by former OpenAI CTO Mira Murati, has released Inkling, a 975-billion parameter mixture-of-experts model with open weights designed for enterprise customization and on-premises deployment. The company explicitly positions the model as focused on customizability rather than benchmark dominance.
Chinese food delivery giant Meituan has open-sourced LongCat-2.0, a 1.6 trillion-parameter AI model that the company claims was trained entirely on Chinese-made chips using a 50,000-chip cluster. The release demonstrates China's growing capability to develop large-scale AI models without relying on Nvidia hardware amid U.S. export restrictions.
SpaceX has signed a multiyear agreement worth up to $6.3 billion with open-source AI startup Reflection, providing access to Nvidia GB300 chips at its Colossus data center for $150 million monthly through 2029. The deal comes as SpaceX enters the bond market for the first time and reports cash reserves exceeding $100 billion.
Chinese AI startup Z.ai (formerly Zhipu AI) has released GLM-5.2, a 753-billion parameter open-weight language model that achieved the highest score (51) among open models on the Artificial Analysis Intelligence Index and reportedly outperforms GPT-5.5 on long-horizon coding benchmarks at one-sixth the cost. The model features a 1-million token context window and is available under MIT license on Hugging Face.
Databricks released Omnigent, an open-source 'meta-harness' under Apache 2.0 license that composes, governs, and shares AI agents across multiple platforms including Claude Code, Codex, and Pi. The project was built in collaboration with Neon.
Chinese AI lab Moonshot AI launched Kimi K2.7 Code, a one-trillion parameter open-weights coding model priced up to 12x cheaper than GPT-5.5 and Claude Opus 4.8, though it trails those models in performance benchmarks and lacks independent verification.
Google has launched Gemma 4 12B, an open-source AI model that can analyze audio and video while running entirely on enterprise laptops with 16GB of RAM, marking a focus on smaller, locally-deployable models rather than larger cloud-based alternatives.
Shanghai-based AI company MiniMax has unveiled its M3 flagship model designed for coding agents and automated workflows, featuring a one-million-token context window. The launch enters the competitive open-source AI coding space, though the company has stopped short of a full open-source commitment.
Blockchain privacy company Aztec Labs has acquired Obsidion, the developer of ZKPassport, and committed to keeping the protocol and its iOS application open-source amid growing pressure for age verification solutions.
Anthropic announced that its Claude Mythos Preview AI model has detected over 23,000 potential security vulnerabilities across more than 1,000 open source software projects, representing a significant automated security audit of the OSS ecosystem.
NanoCo, creator of the open-source AI agent framework NanoClaw, raised $12 million in seed funding and launched its enterprise assistant platform. The announcement follows 250,000 downloads of the NanoClaw framework.
Alibaba Group Holding has released its flagship AI model Qwen3.8-Max with 2.4 trillion parameters, making it accessible via QwenCloud. The company plans to release the model's weights for public download the week of August 10, with benchmarks showing performance approaching Anthropic's Claude Fable 5.
NVIDIA's research lab released cuda-oxide v0.1.0, a custom Rust compiler backend that compiles GPU kernels directly to PTX through a novel compilation pipeline. The experimental tool enables Rust developers to write CUDA kernels natively.
OpenAI launched Multipath Reliable Connection (MRC), a new open networking protocol developed with AMD, Broadcom, Intel, Microsoft, and NVIDIA to improve GPU networking performance in large-scale AI training infrastructure.
NVIDIA has released Ising, the first open quantum AI model family designed for hybrid quantum-classical computing systems, marking a significant step in making quantum computing more accessible for practical applications.
MiniMax released model weights for MiniMax M2.7 on Hugging Face, a self-evolving agent model that scores 56.22% on SWE-Pro and 57.0% on Terminal Bench 2. The company also released MMX-CLI, a command-line interface providing native access to image, video, speech, music, vision, and search capabilities.
US startup Arcee AI invested approximately half of its total venture capital to train Trinity-Large-Thinking, a 400 billion parameter open reasoning model designed to compete with Anthropic's Claude Opus in agent tasks.
Chinese AI company Z.AI unveiled GLM-5.1, a 754B parameter open-source agentic model that achieves state-of-the-art performance on SWE-Bench Pro and can run autonomously for up to 8 hours on long-horizon engineering tasks.
RightNow AI has released AutoKernel, an open-source framework that applies an autonomous agent loop to GPU kernel optimization for arbitrary PyTorch models, aiming to automate one of the most specialized tasks in ML engineering.
Google's DeepMind released Gemma 4, a family of four open-weight AI models ranging from edge devices to workstations, marking the first time Google has used the fully open Apache 2.0 license instead of its previous restricted licensing. The models support up to 256K context windows and 140+ languages.
