building Humanbased (https://xitter.cf/t.co/wztSjtJ6D3) and XnY (https://xitter.cf/t.co/AW6JsT4E29, was Codatta)
San Mateo, CA, USA
Joined May 2009
- Tweets2.8K
- Following1.3K
- Followers1.6K
- Likes2.8K
Yi retweeted
Announcing Artificial Analysis Intelligence Index v5, featuring 2 new benchmarks and 3 upgrades. For the first time, the majority of the Index is built on private datasets
The Artificial Analysis Intelligence Index is our composite measure of model capabilities across tasks reflecting real-world usage today. v5 is our biggest upgrade since v4 launched in January. With v5, 60% of the Index is now based on evaluations with private test sets or held-out answers.
Alongside Intelligence Index v5, we’re launching the Artificial Analysis Progress Index v1, which tracks model capability over time on one continuous scale.
New to the Intelligence Index:
➤ FrontierCode 2.0: Cognition's newly-launched private benchmark of real-world coding tasks, graded on whether a repository maintainer would merge the agent's change
➤ Terminal-Bench-Science: 70 agentic scientific research tasks, run in a terminal environment
Upgraded:
➤ HLE-Diamond (replacing Humanity's Last Exam): a refined 1,000-question version of Humanity's Last Exam, split evenly between reasoning and knowledge questions
➤ AA-Omniscience v1.1: An updated dataset with saturated and public set questions removed, new questions added, and ambiguous or erroneous questions revised
➤ CritPt v2: a private, 40-question hard subset of CritPt v2, revised by the CritPt authors and run with a code-execution tool
Staying in the Index: AA-Briefcase, AutomationBench-AA, GDP.pdf and Terminal-Bench 4.0
Removed from the Index: GDPval-AA v2.1, SciCode, and AA-LCR v1.1
Key results:
➤ Claude Opus 5.5 (max) leads at 58, ahead of GPT-6 Astra (max, 57) and GPT-6.1 Sol (max, 55): Anthropic and OpenAI hold the top five models. Gemini 4 Argon (high) is next at 50
➤ The open weights gap widens from 3 points to 12: Kimi K3.1 Pro (weights not yet released) is the top open weights model at 46. Among released weights, DeepSeek V4.1 Flash 1017 (max) and GLM-5.3 (max) lead at 32 - 26 points behind the leader and below Claude Haiku 5.5 (max, 36)
➤ Cost per task varies by more than 13x among the leading models: of the 10 highest-scoring models shown, GPT-6.1 Terra (max) is the cheapest at $0.96 per Intelligence Index task and Claude Fable 5.1 (max) the most expensive at $13.04
Yi retweeted
"Agentic World Action Models"
What if robots could write their own code to plan actions, while also predicting how the physical world will respond?
NVIDIA's new AW-0 combines three capabilities in one model: generating reusable skill programs, predicting future video frames, and producing robot actions. All three are generated together, which enables code-based planning to guide physical execution.
On RoboTwin 2.0, AW-0 achieves 50.7% average success, compared to 41.3% for π0.5. Across eight real-world manipulation tasks, it scores 71.1% vs 46.0% for the strongest baseline, including tasks like folding towels and solving a Rubik's Cube.
alphaxiv.org/abs/2610.30136
Yi retweeted
Made a skill for dimensional drawings of UI components.
Inspired by Apple's accessory drawings, it documents a component's architecture into a spec sheet.
github.com/enesozturk/dimens…
Yi retweeted
The Decisions API is here!!!
But one of my favorite demos got left on the cutting room floor - an outfit picker!
Yi retweeted
Private. Decentralized. Lens × NymVPN. Explore the drop: nym.lens.xyz/
Yi retweeted
About 2 years ago, we launched a mini project called abtest.design - a curated library of A/B test results from best-in-class apps like Uber, Duolingo, and Headspace.
It was our first attempt at answering questions we constantly ask when browsing references on @mobbin:
“Why did they choose this design pattern?”
“Does this pattern actually perform better?”
We wanted to go one step further and surface the real-world data and reasoning behind those design decisions.
Since launch, the site has amassed 600k+ page views, and it’s been cool seeing folks organically share it with the design community 🫶
We’re planning to add more tests over time, and if you’d like to contribute and help grow the library, DM me or fill out the form below 👇
Yi retweeted
One of my recent favorite AI use cases: asking it to try and break the stuff I built. Adding lots of data, long names, unusual emails, labels, etc. Coming up with worst case scenario basically.
This is a vibe-coded demo, but you get the idea.
Yi retweeted
Official Notice
The GitHub account [github.com/codatta] is impersonating our legacy brand. It is not affiliated with or authorized by us, and the token it promotes has nothing to do with our team.
We remain fully committed to $XNY and have no plans to issue a new token.
Follow our official channels for an important release next month.
Yi retweeted
New skill: /apple-design
Apple’s WWDC videos are a goldmine of knowledge. I’ve combed through my favorite ones and came up with 17 design and motion principles.
Use them to review existing work or when working on something new to get it right.
github.com/emilkowalski/skil…
Yi retweeted
I’ve been tinkering with some new skills for my skills repo over the weekend, and it seems like there’s so much untapped potential for design engineering skills.
/emil-design-eng skill can already help you a ton, and it’s doing more than 100k installs already, but there’s more to come soon.
I plan to add more fine-tuned skills for specific use cases, but since AI produces non-deterministic results, there’s still quite a bit of testing to do as I want the answers to be precise and correct.
In the meantime, you can try the two existing skills:
github.com/emilkowalski/skil…
Yi retweeted
I think so many concepts in design need to really be felt and played with, not just shown.
That's why Interface Craft is full of interactive figures that demonstrate topics.
I always look forward to making them; this is one of my favorites!
Means & Methods is now live on interfacecraft.dev.
A collection of practical techniques to achieve excellence in interface design, covering 100+ topics across 11 chapters, with plenty of interactive examples and code.
New members welcome; enjoy!
Yi retweeted
cool internet things pt. 6
profconradi.com - math visualized
music.jessyin.world - visual instruments
everynoise.com - every music genre ever
poetry.camera - a camera that prints poems
marijanapav.com/stamps - digital stamp collection
liumichelle.com/sketchbook - a living digital sketchbook
some of the cool people behind these:
@michelletliu, @marijanapav, @rpavlini, @S_Conradi, @EveryNoise, @S_Conradi, @itsjessyin, @kelin_online
cool internet things pt. 5
digibouquet.com - digital flowers
xrageroom.com - destroy any tweet
yeguessr.com - guess the kanye album
flower-buds.vercel.app - friendship tracker
counter-print.co.uk - beautiful design books
nycstoops.com - nyc building analysis + ratings
presentandcorrect.com - beautiful stationery + desk stuff
some of the cool people behind these: @pau_wee_, @immike_wing, @downloadlos, @marcgmbh, @maggiexgao, @jen__jpeg, @Counterprint
Yi retweeted
It’s never been easier to design your dream house.
Draw a shape. Define your rooms. Set your constraints.
@DraftedAI generates complete floor plans, elevations, and 3D home designs in seconds.
Over the last month, 120,000 people generated 325,000+ home designs with Drafted.ai.
Yi retweeted
Launching our new paper on arXiv: we trained the largest multilingual food model ever built.
4.1M recipes. 7 languages. 1,790 ingredients. 300 dimensions.
All of human cooking compressed into 2 megabytes.
