-
Anthropic’s Claude Science ships every figure with the code that made it
Anthropic dropped its biggest product of the day on June 30: Claude Science, a desktop AI workbench for scientists. Not a paper, not a model — an actual macOS/Linux app. It hit 391 points on Hacker News and got picked up across tech and pharma press. What it actually is Think of it as a… Continue reading
-
Google Nano Banana 2 Lite (Gemini 3.1 Flash-Lite Image): a picture in 4 seconds for $0.034 per 1,000
Google DeepMind just shipped the cheapest, fastest member of the Nano Banana family. Model ID: gemini-3.1-flash-lite-image. It’s an image-generation model built for one thing — throughput. One image in under 4 seconds, roughly 2.7× faster than Gemini 3.1 Flash Image, at a flat $0.034 per 1,000 images. On Artificial Analysis’s text-to-image leaderboard it sits at… Continue reading
-
Claude Code is steganographically marking requests — a researcher decoded how
Claude Code is Anthropic’s coding agent — the CLI tool that lives in your terminal, reads your files, and runs shell commands to write and fix code. A privacy researcher pulled apart its binary and found something nobody asked for: it quietly stamps hidden markers into the system prompt on every affected request. What the… Continue reading
-
Claude Sonnet 5 scores 63.2% on SWE-bench Pro at a third of Opus 4.8’s price
Anthropic shipped Claude Sonnet 5 on June 30, and the pitch is blunt: this is now the cheapest way to run agents that don’t fall apart. It’s a mid-tier LLM built for agentic workflows — planning, calling tools, driving a browser or terminal, and grinding through long coding tasks without a human babysitting every step.… Continue reading
-
agency-agents by msitarzewski: 232 specialist agents that turn Claude Code into a full company
A Reddit thread about agent specialization turned into the fastest-climbing repo of the week. msitarzewski’s agency-agents hit 120K GitHub stars, adding nearly 1,800 in a single day. It’s not another prompt pack. What it actually is It’s an open-source roster of 232 specialized agents spread across 16 departments — frontend engineers, backend API experts, security… Continue reading
-
v0 Design Systems 2.0 (Vercel) teaches AI to build UI with your real components
Every AI UI generator has the same chronic problem: the output looks fine, but it’s not yours. Wrong buttons, wrong spacing, none of your team’s components. v0 Design Systems 2.0 from Vercel goes straight at that. Import your system once, generate on-brand forever It’s a coding tool. You point v0 at wherever your design system… Continue reading
-
Cursor for iOS turns your phone into a coding-agent dispatch console
Cursor shipped its first native iPhone and iPad app on June 29, in public beta for paying users. It is not a code editor squeezed onto a small screen. It is a remote control for AI agents. What it actually does Pick a repo, pick a model, then describe the task by typing or talking.… Continue reading
-
Meta Brain2Qwerty v2 jumps non-invasive brain-to-text from 8% to 61% word accuracy
Meta FAIR just made reading minds a software problem instead of a surgery problem. Brain2Qwerty v2 is a brain-to-text model that decodes the sentences you silently type — straight from brain activity, no implant, no opening the skull. You wear a MEG (magnetoencephalography) helmet, think the words, and the model spits out the text. What… Continue reading
-
Meituan’s LongCat-2.0 is a 1.6T open-weight model trained on 50,000 Chinese ASICs — no NVIDIA
The food-delivery company just dropped a frontier-scale LLM. LongCat-2.0 is a Mixture-of-Experts model with 1.6 trillion total parameters, ~48B activated per token, and a 1M-token context window. The whole pretraining run — 35T+ tokens — happened on 50,000+ domestic AI ASICs in superpod clusters. NVIDIA wasn’t in the building. What it actually is This is… Continue reading
-
Ornith-1.0 (DeepReinforce self-scaffolding coding models): open weights that write their own RL scaffold, 397B hits 82.4 on SWE-bench
DeepReinforce just open-sourced Ornith-1.0, a family of agentic coding models running from a 9B Dense for edge boxes up to a 397B MoE. Built on Gemma 4 and Qwen 3.5, MIT-licensed, all weights already on Hugging Face. It hit the HN front page fast. The model that builds its own scaffold Everyone else hand-writes the… Continue reading
Top AI Product
Every day, hundreds of new AI tools launch across Product Hunt, Hacker News, and GitHub. We dig through the noise so you don't have to — surfacing only the ones worth your attention with honest, no-fluff reviews. Explore our latest picks, deep dives, and curated collections to find your next favorite AI tool.
Top AI Product 