Every AI News story tagged with both AI Research and AI Tools — the two topics side by side, updated as new articles publish.
24 articles
PrismML has released Bonsai 27B, a 27-billion-parameter open-weight model compressed to under 4GB, small enough to run natively on a phone while keeping most of its reasoning ability.
Read more →Mira Murati's startup released Inkling, a 975-billion-parameter open-weight model built for developers to fine-tune and adapt, rather than to top leaderboards.
Read more →
DeepSeek is the Chinese AI lab whose cheaply trained, open-weight models triggered Nvidia's biggest one-day stock drop and reshaped how the industry talks about AI costs.
Read more →
Generative AI creates new text, images, audio, and code by learning patterns in data, instead of just sorting or predicting like older AI systems. Here's how it actually works.
Read more →An internal audit found that roughly a third of SWE-Bench Pro's coding tasks are flawed, prompting OpenAI to withdraw its recommendation that developers rely on the benchmark.
Read more →GLM is the open-weight AI model family from Chinese lab Zhipu AI (Z.ai) — free to download and self-host, and competitive with top US models at a fraction of the API cost.
Read more →Chatbot Arena (now Arena) is a free platform that ranks AI models through blind human votes — and it’s the leaderboard the whole AI industry watches.
Read more →Chinese tech company Meituan has disclosed that "Owl Alpha," an anonymous model that topped OpenRouter rankings for two months, was its 1.6-trillion-parameter LongCat-2.0 — trained entirely on domestic chips and now free under an MIT license.
Read more →Mistral's new open-source model for Lean 4 fully saturates the miniF2F theorem benchmark, solves 87% of graduate-level math tests, and found five previously unknown bugs in production code — at roughly $4 per proof versus $300 for competing systems.
Read more →Anthropic released Claude Sonnet 5 on June 30, delivering near-Opus-level reasoning and agentic capabilities at introductory API pricing of $2 per million input tokens.
Read more →Anthropic released Claude Science in beta on June 30 — an AI workbench connecting scientists to 60-plus biological databases with reproducible outputs and built-in compute management, available to all paid subscribers.
Read more →An AI benchmark is a standardized test used to compare how capable different AI models are. Here's what the major ones measure — and why the scores don't always tell the full story.
Read more →
Google has released Gemini Omni Flash in public preview, a multimodal AI model that generates and edits short videos through natural-language conversation, available via the Gemini API at $0.10 per second of output.
Read more →Chinese technology company Meituan has released LongCat-2.0, a 1.6-trillion-parameter coding model trained entirely on domestic Chinese chips, under the MIT License — the first frontier-class open-source model built without Western semiconductors.
Read more →Open-weight AI models release their trained parameters publicly so anyone can download, run, or fine-tune them — unlike closed models such as GPT-4 or Gemini, which are accessible only through paid APIs. Here is what that difference means in practice.
Read more →Anthropic's new Claude Science workbench integrates over 60 scientific databases and tools into a unified AI environment for life sciences researchers, launching in beta on June 30, 2026.
Read more →AI reasoning models think through problems step-by-step before answering, making them far more accurate on math, coding, and complex tasks — but slower and costlier than standard models.
Read more →AI is transforming how scientists search literature, analyze data, and write up results. Here is what the tools can do today and where to start.
Read more →Anthropic released Claude Science in beta on June 30, an integrated environment connecting AI to 60+ scientific databases and lab compute for biomedical and life sciences research.
Read more →Claude is an AI assistant developed by Anthropic, built on a safety-first training approach and available through the web, mobile apps, and a developer API.
Read more →Open-source AI models release their trained weights publicly, letting anyone run, study, or customize them without paying API fees. This guide explains what that means in practice and how to get started.
Read more →