HN user
beklein.com
Details about the methods can be found here: https://artificialanalysis.ai/methodology/intelligence-bench...
Specifically they use this harness: https://github.com/ArtificialAnalysis/Stirrup
Some other (as in better) sources I found:
- https://huggingface.co/spaces/Soofi-Project/Pretraining-Tech...
Thanks for sharing the paper, it seems my intuition on size was bad but also my intuition on max satellite density for LEO was wrong.
To get a better sense of the scale, if you are viewing this app on a 4K display, with the planet measuring about 2,000 pixels across Earth’s diameter is approximately 12,742 km (7,918 miles), so each pixel represents about 6.37 km (3.96 miles).
A Starlink satellite is roughly 6 m (20 ft) wide without its solar panels. This means a one-pixel satellite marker is shown at roughly 1,000 times its true size. So even if this image already looks extremely crowded, the dots are still massively exaggerated. Visually, there would be roughly another factor of 1,000 before the satellites themselves were shown at their true scale—although this does not mean that orbit could easily accommodate 1,000 times more satellites but I guess there is still some space in space.
All AI labs really need to stop using truncated y-axes for benchmark bar charts...
https://mistral.ai/_astro/cm-engish_ZhlvoT.webp?dpl=6a3a94bd...
A good visual explanation of how text diffusion models like DiffusionGemma work: https://newsletter.maartengrootendorst.com/p/a-visual-guide-...
Principles of programming:
1. Break things down into small units
2. Think about sequence
3. Find patterns
4. Focus on the important things
5. Visualize sequences in your mind
Love the silly music and the way they teach, thanks for sharing this!
Also relevant to this is the newest episode of The Lightcone Podcast with Quan Vuong, co-founder of PI and, one of many co-authors of that paper.
Thanks for sharing this here, it's a beautiful video. The images are linked in the description but if anybody is reading this, check: https://www.flickr.com/photos/nasa2explore/
"... the first early version of Claude Mythos Preview was made available for internal use on February 24. In our testing, Claude Mythos Preview demonstrated a striking leap in cyber capabilities relative to prior models, including the ability to autonomously discover and exploit zero-day vulnerabilities in major operating systems and web browsers."
More infos here: https://red.anthropic.com/2026/mythos-preview/
As a big Codex user, with many smaller requests, this one is the highlight: "In Codex, GPT‑5.4 mini is available across the Codex app, CLI, IDE extension and web. It uses only 30% of the GPT‑5.4 quota, letting developers quickly handle simpler coding tasks in Codex for about one-third the cost." + Subagents support will be huge.
Not sure why you think Anthropic has not the same problems? Their version numbers across different model lines jump around too... for Opus we have 4.6, 4.5, 4.1 then we have Sonnet at 4.6, 4.5, and 4.1? No version 4.1 here, and there is Haiku, no 4.6, but 4.5 and no 4.1, no 4 but then we only have old 3.5...
Also their pricing based on 5m/1h cache hits, cash read hits, additional charges for US inference (but only for Opus 4.6 I guess) and optional features such as more context and faster speed for some random multiplier is also complex and actually quiet similar to OpenAI's pricing scheme.
To me it looks like everybody has similar problems and solutions for the same kinds of problems and they just try their best to offer different products and services to their customers.
Perhaps useful, I discovered: https://github.com/agent-infra/sandbox
All-in-One Sandbox for AI Agents that combines Browser, Shell, File, MCP and VSCode Server in a single Docker container.
Some more info here: https://developers.openai.com/api/docs/models/gpt-realtime-1...
- $4 input, $0.4 cached input, $16 output
- 32,000 context window
- 4,096 max output tokens
- Sep 30, 2024 knowledge cutoff
Love the models, speed, and capabilities. Just sad that they are not getting the publicity and adoption right now, but hopefully in the future.
Sound on!
Song name is: Windowdipper from ꪖꪶꪶ ꪮꪀ ꪗꪖꪶꪶ by Jib Kidder
The end result would be a normal PPT presentation, check https://sli.dev as an easy start, ask Codex/Claude/... to generate the slides using that framework with data from something.md. The interesting part here is generating these otherwise boring slide decks not with PowerPoint itself but with AI coding agents and a master slides, AGENTS.md context. I’ll be showing this to a small group (normally members only) at IPAI in Heilbronn, Germany on 03/03. If you’re in the area and would like to join, feel free to send me a message I will squeeze you in.
Not my normal use-case, but you can always fall back and ask the AI coding agent to generate the diagram as SVG, for blocky but more complex content like your examples it will work well and still is 100% text based, so the AI coding agents or you manually can fix/adjust any issues. An image generation skill is a valid fallback, but in my opinion it's hard to change details (json style image creation prompts are possible but hard to do right) and you won't see changes nicely in the git history. In your use case you can ask the AI coding agent to run a script.js to get the newest dates for the project from a page/API, then it should only update the dates in the roadmap.svg file on slide x with the new data. This way you will automagically have the newest numbers and can track everything within git in one prompt. Save this as a rule in AGENTS.md and run this every month to update your slides with one prompt.
In my AGENTS.md file i have a _rule_ that tells the model to use Apache ECharts, the data comes from the prompt and normally .csv/.json files. Prompt would be like: "After slide 3 add a new content slide that shows a bar chart with data from @data/somefile.csv" ... works great and these charts can be even interactive.
I love this! I use coding agents to generate web-based slide decks where “master slides” are just components, and we already have rules + assets to enforce corporate identity. With content + prompts, it’s straightforward to generate a clean, predefined presentation. What I’d really want on top is an “improv mode”: during the talk, I can branch off based on audience questions or small wording changes, and the system proposes (say) 3 candidate next slides in real time. I pick one, present it, then smoothly merge back into the main deck. Example: if I mention a recent news article / study / paper, it automatically generates a slide that includes a screenshot + a QR code link to the source, then routes me back to the original storyline. With realtime voice + realtime code generation, this could turn the boring old presenter view into something genuinely useful.
Honest question: would a normal CS student, junior, senior, or expert software developer be able to build this kind of project, and in what amount of time?
I am pretty sure everybody agrees that this result is somewhere between slop code that barely works and the pinnacle of AI-assisted compiler technology. But discussions should not be held from the extreme points. Instead, I am looking for a realistic estimation from the HN community about where to place these results in a human context. Since I have no experience with compilers, I would welcome any of your opinions.
Relevant blog post from simonw: https://simonwillison.net/2026/Feb/7/software-factory/
https://www.spacex.com/updates#xai-joins-spacex additionally the longer article on SpaceX site
This will actually work well with my current workflow: dictation for prompts, parallel execution, and working on multiple bigger and smaller projects so waiting times while Codex is coding are fully utilized, plus easy commits with auto commit messages. Wow, thank you for this. Since skills are now first class tools, I will give it a try and see what I can accomplish with them.
I know/hope some OpenAI people are lurking in the comments and perhaps they will implement this, or at least consider it, but I would love to be able to use @ to add files via voice input as if I had typed it. So when I say "change the thingy at route slash to slash somewhere slash page dot tsx", I will get the same prompt as if I had typed it on my keyboard, including the file pill UI element shown in the input box. Same for slash commands. Voice is a great input modality, please make it a first class input. You are 90% there, this way I don't need my dictation app (Handy, highly recommended) anymore.
Also, I see myself using the built in console often to ls, cat, and rg to still follow old patterns, and I would love to pin the console to a specific side of the screen instead of having it at the bottom and pls support terminal tabs or I need to learn tmux.
Thank you! Learned something new today. Will try to look out for this trick on other Lovable sites I will stumble upon.
Would love to see the original prompt for Nano Banana from OP somewhere. One that yields decent results, for me, is:
{ "image_generation_prompt": { "subject_focus": { "primary": "Architectural exterior scene", "constraint": "Strictly preserve original building geometry, facade details, and structural layout", "reference_adherence": "High structural fidelity to input image" }, "environment_and_season": { "season": "Late November, very late autumn", "weather": "Post-rain, overcast, gloomy, high humidity", "sky": "Heavy grey cloud cover, diffuse white/grey light, no direct sunlight", "ground_texture": "Wet asphalt/pavement, highly reflective puddles, wet concrete, scattering of wet brown decaying leaves" }, "vegetation_details": { "trees": "Leafless branches, dormant skeletal trees, sparse lingering brown foliage", "color_palette": "Desaturated greens, browns, greys, russet, damp earth tones", "state": "Winter-ready, wet bark, dormant landscaping" }, "human_element": { "density": "Sparse, minimal crowd", "clothing": "Heavy winter coats, scarves, boots, muted colors", "activity": "Walking briskly to avoid cold, holding closed wet umbrellas, hurrying, heads down against the wind", "mood": "Solitary, cold, urban transit" }, "photographic_style": { "medium": "Realistic architectural photography", "camera": "35mm lens, sharp focus on architecture", "tone": "Cinematic, moody, desaturated, cool color temperature, blue-grey tint", "quality": "8k resolution, high dynamic range, hyper-realistic textures" } } }
As far as I can tell, they say: "Mission control and data distribution are managed by EUMETSAT." They have published their own blog post here: https://www.eumetsat.int/features/see-earths-atmosphere-neve...
There they say that: "Observations made by MTG-S1 will feed into data products that support national weather services …". So I guess there will be no simple, publicly available REST API or so... but if anybody finds anything, let us know here :)
Wasn't aware you're so active on HN; sorry for stealing your karma.
Love the idea of a dedicated series/pod where normal people take on hard problems by using and leveraging the emergent capabilities of frontier AI systems.
Anyway, thanks for pod!
The Latent Space podcast just released a relevant episode today where they interviewed Kevin Weil and Victor Powell from, now, OpenAI, with some demos, background and context, and a Q&A. The YouTube link is here: https://www.youtube.com/watch?v=W2cBTVr8nxU
I wonder if electrical engineers felt this way about the reliability of the silicon crystal lattice their circuits rely upon…