open-weight models

Alibaba has previewed Qwen3.8-Max-Preview, a 2.4 trillion-parameter multimodal AI model that it says ranks behind only Anthropic’s Claude Fable 5 — but the claim arrives without the benchmark data, model card, or independent testing that enterprise developers would need to treat it as more than...

Moonshot AI’s Kimi K3 is not yet the outright best AI model in the world, and Moonshot itself concedes that it trails Anthropic’s Claude Fable 5 and OpenAI’s GPT-5.6 Sol in overall performance. But its July 16 release has changed the more important calculation: a 2.8-trillion-parameter Chinese...

GitHub made Moonshot AI’s Kimi K2.7 Code generally available in the GitHub Copilot model picker on July 1, 2026, giving Copilot users access to a Beijing-built open-weight coding model hosted through Microsoft Azure rather than Moonshot’s own infrastructure. That is not just another model-card...

GitHub made Kimi K2.7 Code generally available in GitHub Copilot on July 1, 2026, adding the open-weight coding model to Copilot’s model picker first for Pro, Pro+, and Max users, with Business and Enterprise access planned over the coming weeks. The announcement is easy to read as just another...

Microsoft CEO Satya Nadella said in an interview released Friday that companies should build AI systems shaped by their own data, workflows, and institutional memory rather than depending entirely on a small group of frontier model providers. His argument is not anti-OpenAI, despite Microsoft’s...

Canonical is preparing to make AI a first-class part of Ubuntu, but its pitch is deliberately different from the Copilot-heavy strategy that has defined Microsoft’s recent Windows roadmap. The company says Ubuntu’s AI features will arrive gradually over the next year, with an emphasis on local...

The mainstream AI assistants sweeping into everyday workflows are now being built and operated by companies whose core businesses are, in many cases, advertising and attention monetization — and that structural fact reshapes every privacy, legal, and architectural decision engineers and IT...

Microsoft researchers have shown that a single, seemingly benign unlabeled prompt can erase safety guardrails in a wide range of modern open-weight models — a finding that forces a hard rethinking of how enterprises and vendors evaluate alignment, fine-tuning workflows, and the threat model for...

Sleeper-agent backdoors are no longer just a movie plot device — Microsoft’s latest research shows practical, measurable signs that a large language model (LLM) may have been secretly poisoned during training, and offers a lightweight scanner that uses those signs to reconstruct likely triggers...

Microsoft and Mistral have quietly lowered the barrier between open-weight frontier models and enterprise production: Mistral Large 3 is now a first‑party model in Microsoft Foundry on Azure, offered under Apache 2.0 terms and packaged with the governance, observability, and agent tooling...

Cisco’s latest security sweep has found that many of the most widely used open-weight large language models are alarmingly easy to manipulate with a small series of crafted prompts — and multi-turn (conversation) attacks are the most effective vector, producing success rates two to ten times...

Microsoft’s newly announced multi-year partnership with Mistral AI marks one of the boldest moves yet in the cloud–AI arms race: Azure will provide supercomputing capacity, developer tooling, and commercial distribution channels for Mistral’s flagship models, while Mistral supplies both...

Cloud providers’ quiet September previews revealed a pivot: enterprises are no longer satisfied with raw model accuracy alone — they want platforms that deliver security boundaries, governance, and predictable operations so generative AI can safely move into production. Background / Overview...

September’s quiet preview windows at the major cloud providers are shaping up to be one of the clearest signals yet that enterprise AI is moving from model-first experimentation into regulated, operational production—and the changes being previewed are less about raw model accuracy and more...

Cloud providers’ recent September preview releases from Microsoft, Amazon Web Services, and Google aren’t incremental feature drops — they’re a clear signal that enterprise expectations for cloud AI have shifted from “which model is best?” to “which platform makes models secure, auditable, and...

Cloud providers’ September previews are not incremental checkbox updates; they are a clear signal that enterprises expect AI clouds to be more than high‑performance models — they must be secure, auditable, and operationally mature enough to run production workloads at scale. Background...

Ollama’s latest Windows 11 GUI makes running local LLMs far more accessible, but the single biggest lever for speed on a typical desktop is not a faster GPU driver or a hidden setting — it’s the model’s context length. Shortening the context window from tens of thousands of tokens to a few...

Microsoft has unveiled a significant expansion of its artificial intelligence ecosystem by introducing OpenAI’s open-weight language models, gpt-oss, to both Azure AI Foundry and Windows AI Foundry. This move marks a pivotal shift toward openness, flexibility, and developer empowerment...

For the first time, OpenAI’s artificial intelligence models are now available on a cloud computing platform outside of Microsoft Azure, marking a significant milestone in the competitive landscape of enterprise AI. Amazon Web Services (AWS) has announced that it will offer OpenAI’s new...

OpenAI's recent release of the GPT-OSS models, gpt-oss-120b and gpt-oss-20b, marks a significant milestone in the democratization of artificial intelligence. These open-weight models, available under the Apache 2.0 license, offer developers unprecedented flexibility and control over AI...