event
Gemini 4 Release
-
TL;DR: We pitted Claude Opus 5.5#xhigh (first picture) against Gemini 4 Hydrogen#xhigh (second picture) in Pochade, an internal physical oil-painting engine we have been building (still unreleased, planning to open it up once we iron out t…
-
Get Started with Gemini 4: Explicit Context Caching Not Supported (docs.cloud.google.com via hn) tool-usegeminiagentic 1d
The Gemini 4 model family includes the following models: Gemini 4 Hydrogen ( gemini-4-hydrogen ): Built for speed and efficiency across everyday coding and agentic workflows, multimodal analysis, and tool use.Gemini 4 Carbon ( gemini-4-car…
-
Anthropic released Fable 5.2 and Mythos 5.2 on Monday. Fable is the public version at $8/$40 per million tokens, while Mythos is the same underlying model restricted to vetted cybersecurity and life-science organizations.
-
Gemini 4 Hydrogen (High) Intelligence, Performance and Price Analysis (artificialanalysis.ai via hn) gemini 3d
Gemini 4 Hydrogen (High) Intelligence, Performance & Price Analysis Model summary Gemini 4 Hydrogen (High) is above average in intelligence and reasonably priced when comparing to other models of similar price. It's also quicker than avera…
-
llm-gemini 0.37 (simonwillison.net) gemini 3d
12th November 2026 Google launched Gemini 4 Hydrogen today - the lowest cost member of the Gemini 4 family. It's actually not too far off the price of its much bigger siblings, Carbon and Argon.
-
Google rolls out Gemini 4 Hydrogen, finally replacing 3.8 Flash (arstechnica.com) gemini 3d
Google is continuing its elemental AI naming scheme with the new Gemini 4 Hydrogen. These names are less descriptive than the Pro/Flash monikers of earlier releases, but Hydrogen’s purpose is clear: It replaces the faster, more economical…
-
Gemini 4 Hydrogen and Orchestrator (antigravity.google via hn) gemini 3d
Explore our next generation products Built for developers and teams in the agent-first era Everything you need to stay up-to-date and get help Gemini 4 Hydrogen and Orchestrator Today, we’re rolling out Gemini 4 Hydrogen inside Google Anti…
-
Google Search gets a major upgrade with Gemini 4 Hydrogen (blog.google via hn) gemini 3d
Google Search gets a major upgrade with Gemini 4 Hydrogen Listen to article This content is generated by Google AI. Generative AI is experimental Today, Google Search is bringing cutting-edge intelligence to everyone with Gemini 4 Hydrogen.
-
I've been using Codex extensively for development, currently on the $100/month plan. I wanted to find out whether a cheaper AI coding environment could deliver comparable results without constantly running into usage limits.
-
Tell HN: Gemini Hydrogen Reachable via Vertex (news.ycombinator.com) gemini 4d
Currently on the eu and global endpoint gemini-4-hydrogen is available.
-
Fable vs Carbon vs Astra - The Robotic Dog Showdown (www.reddit.comhttps) geminicodex 4d
One prompt, one skill, create a robotic dog, rigged and animated. Xhigh/high agent does research, and a low agent builds the 3D model.
-
same price per token, 3-4.5x the bill: gemini 4 carbon vs argon at default thinking (www.reddit.com via reddit) gemini 4d
gemini 4 carbon opened to paid api users yesterday, and carbon and argon list the same price ($2 in / $10 out per million until nov 30). so i ran both through the api on the same tasks to see what it costs in practice.
-
Fable 5.2 feels unusable? (www.reddit.com via reddit) hallucinationgemini 5d
My own tests with Fable 5.2 Max vs Gemini 4 Carbon Fable 5.2 Max seems to respond fast without thorough thinking, makes writing mistakes (exaggerations), and hallucinates much more. I recently just found out on https://artificialanalysis.a…
-
Gemini 4 Carbon now available (twitter.com via hn) geminiagentic 5d
Our strongest frontier model, Gemini 4 Carbon, is now available. Built for deeply complex, long-horizon workflows that push a frontier model’s intelligence, Carbon delivers frontier performance in agentic workflows across complex knowledge…
-
Gemini 4 Carbon Model Card [pdf] (storage.googleapis.com via hn) gemini 5d
could not extract summary
-
Gemini 4 Carbon is rolling out today (twitter.com via hn) gemini 5d
Gemini 4 Carbon is rolling out today. It’s built for sustained, long-horizon tasks, reasoning deeply through complex software engineering, cybersecurity, and scientific discovery.
-
Tell HN: Gemini Carbon released (blog.google via hn) gemini 6d
Gemini 4 Carbon: our next frontier model Today, we’re announcing the next frontier model: Gemini 4 Carbon, our strongest model with state-of-the-art cybersecurity and long horizon capabilities. Carbon ranks #1 on the recently updated v5.0…
-
I'm the author of noracu, a free, open-source (MIT) local MCP server (noracu on npm) that lets Claude operate macOS apps and Playwright by element_index, with an experimental iPhone layer through Apple's iPhone Mirroring. The bug: on relea…
-
Work Gemini api key. Alternatives to cursor? (www.reddit.com via reddit) geminicursor 8d
I'll bulletpoint for ease, - I use cursor daily, on ultimate plan. - I have a gemini api key through work that hasn't really got a token limit - works fine for all my projects at work for gemini 4 - I can't use the gemini api key in Cursor…
-
SketchAgentBench – evaluate LLMs on 3D spatial reasoning (www.reddit.com via reddit) gemini 9d
I created this benchmark using the 3D coding playground I previously shared here, and ran 11 models on 25 photos: https://sketchagent.dev https://preview.redd.it/wsodwqz4e20i1.png?width=2562&format=png&auto=webp&s=1ceb0d0f7795bf5607f14c126…
-
Modernising 40-year-old scientific code with Gemini 4 Argon (antigravity.google via hn) gemini 9d
Explore our next generation products Built for developers and teams in the agent-first era Everything you need to stay up-to-date and get help Modernizing 40-year-old scientific code with Gemini 4 Argon Using Gemini 4 Argon in Google Antig…
-
Gemini 4 often hides its failures when searching antibody space (AbMaze) (valenlabs.com via hn) gemini 11d
AbMaze: Evaluating frontier AI agents on protein fitness landscapes Dominik Scherm · November 4, 2026 TL;DR: We built AbMaze, a benchmark that runs an AI agent through an antibody campaign on a fully measured fitness landscape, 65,536 lab-…