Jidé ✨ retweeted
We keep improving performance across Claude, and you may have noticed it starts responding sooner!
Claude Code now gzips the conversation it sends on every turn, so it reaches the model faster. Cloud sessions also start up as soon as you start typing.
For a bit more than two years I’ve been building Ludiz, a tool to make serious games for companies. Onboarding, safety, compliance training.
I’ve worked on it relentlessly. Head down, brick by brick. Every screen redone until it was right, the editor rebuilt from scratch, the way games look reworked again and again until a new game looks great out of the box. On top of that, an assistant that drafts a full game from a simple description, and an editor that lets you take over every detail after.
It took a lot of stubbornness. But it’s there now. When I build a game start to finish today, it’s smooth, it’s clean, it’s the product I wanted to make. And honestly, I’m really proud of it.
Here’s a raw screen recording: I ask for an onboarding game, the assistant builds it, I tweak a title, change the colours, then play a few steps. Waiting times are sped up, everything else is real.
Jidé ✨ retweeted
Just hit #4 on my own product 😭
If you have a product you want to launch, get free backlinks, and put it in front of more users, list it on RankCert.
More updates coming soon 👀
rankcert.com
Jidé ✨ retweeted
We’re rolling out Gemini 4 Argon in Antigravity to paid Google AI Pro subscribers today. Demand has been incredible and we've scaled capacity deliberately to get here. We’ve rolled out a separate weekly-only Argon quota for Pro on top of the existing Gemini bucket for 3.8 Flash so you have the best of Gemini for all your tasks.
Jidé ✨ retweeted
Neither Cartier-Bresson, Banksy nor Cold War Steve could have come up with such an image. Absolutely perfect composition and lighting (even the road signs are perfect)! 😍
Jidé ✨ retweeted
Today we’re releasing Muse Spark Open, the most capable US open-weight model to date.
Built for coding and agentic work, it excels at long-horizon tasks, complex software engineering, and multi-step reasoning.
Key Model Specs:
→ 816B total / 42B active parameters
→ 1M token context window
→ Natively multimodal across text, image, video, audio
Jidé ✨ retweeted
describe a company game and play it in minutes is a fun one. pinned you in france, dofollow's live. let's connect rankcert.com/product/ludiz
Jidé ✨ retweeted
FLUX 3 Image Klein is here.
Open Weights with the quality of a large frontier model. Customizable and fast.
5.6B parameters and runs on a single consumer GPU.
Native 4K, bounding boxes, and pixel-perfect editing.
Jidé ✨ retweeted
FLUX 3 Video is now up to 2x faster and costs 50% less for generating in native 2K and 4K.
We made optimizations so it uses less compute to serve. The new pricing and speed improvements reflect that.
Jidé ✨ retweeted
This is also interesting to think about
Right now it still takes, let's say, 30 minutes to a few hours to build something usable with AI and then you still polish for days/weeks/months
If we assume this will keep getting faster, at some point an AI model will take a second or less to generate the whole thing
Imagine a future where you can create anything in less than second?
How will that change what we choose to create? Will humans even create, or will it all be AI agents coming up with ideas themselves?
The next constraints then become physical? If AI can generate for example hardware plans so fast, it will still take 30 minutes to 3d print that, so maybe the next speed and intelligence optimizations will then start happening in hardware, being able to generate an idea and producing a physical object in a minute or so
I have no idea but very interesting, it's kinda like those real-time video models we saw last month (like my Infinite Slop project), how will content change if you can generate so fast?
🤔
Very creative ad :)
A video editor with no timeline, no tracks and no keyframes.
Takes records camera and screen in one button, the teleprompter follows your voice, and Claude Code does the cut on your own plan.
Free and MIT!
producthunt.com/products/tak…
Jidé ✨ retweeted
New on our Engineering blog: Creating in-house source control.
In May, we merged around 2,500 code changes per day. In September, we merged more than 30,000 per day. As a result, we became bottlenecked by developer infrastructure built for human (rather than agent) scale.
Here, we share how we built a new version control system—and fully migrated our monorepo from git—in five months: anthropic.com/engineering/re…
Fascinating
Over the last few months, we’ve made significant progress in reducing the effectiveness of distillation attacks against Claude.
In this post, we explain how: anthropic.com/research/disti…
