Yes! In Cursor it's particularly great too now they fixed the tool call convention.
AI
-
Codex user growth projected to reach 10 million by October
By
–
Codex user growth, if this trend continues, it would be 10m by October
-
Masayoshi Son invests 75 billion in France for AI datacenter
By
–
Masayoshi Son, Japan's second richest man and SoftBank CEO, is about to invest 75 billion in France to build one of Europe's largest AI datacenters. Why France and not elsewhere? A simple reason: our nuclear electricity, abundant and carbon-free. AI
-
Official DiffusionBlocks Implementation for ViT Image Classification
By
–
Our official implementation of DiffusionBlocks on image classification using Vision Transformers (ViT)
-
Difference in harness between Codex and Claude Code, AI photo
By
–
I talked about the difference in harness between Codex and Claude Code all evening with @lenasituations. (Not at all) Regarding the second photo, I disclaim all responsibility, it's AI.
-
AI becomes the floor, thinking without it is the new advantage
By
–
Everyone now has access to AI. The tool is no longer an advantage: it has become the floor. And as equipment levels out, the differentiator shifts: it will not be in the ability to use these tools, but in that of thinking without them when the stakes
-
Using red grid lines to help models navigate screenshots
By
–
For example, models could not really “see” well. So one thing I used to do before injecting a computer screenshot into the model was overlay red grid lines on top, so the model could see the sections better and know where to click. That’s why this was able to work.
-
Nostalgia for early 2024 AI pipelines and progress
By
–
In the early days you had to build so many things just to make this work.
— Pietro Schirano (@skirano) 31 mai 2026
Computer use wasn’t a thing, models were slow, and you had to pipeline so many models. I do miss those times but it’s amazing to see how far we’ve come.
This was 2024.
https://t.co/Gd8NI4b5vB https://t.co/WB7f7ivZxiIn the early days you had to build so many things just to make this work. Computer use wasn’t a thing, models were slow, and you had to pipeline so many models. I do miss those times but it’s amazing to see how far we’ve come. This was 2024. https://
x.com/skirano/status
/1804219134999192048?s=20
… -
Pro plan’s daily video limit and continuity issues
By
–
Now here's what I'm not going to do: pretend this was flawless. The Pro plan currently caps video generations at 3 per day. When prompting a continuation clip, the model changed character outfits and altered the first character's appearance mid-conversation. Whether that's a
-
Omni’s Editing Advantage: Conversational Video AI, Not Just Generation
By
–
Omni's real advantage isn't output quality. Not yet.
— God of Prompt (@godofprompt) 31 mai 2026
It's the editing model.
Every other video AI tool works the same way: generate, don't like it, start over from scratch. Omni lets you edit the video through conversation. Change the camera angle in step 3, and it still… pic.twitter.com/wka8rI4iiHOmni's real advantage isn't output quality. Not yet. It's the editing model. Every other video AI tool works the same way: generate, don't like it, start over from scratch. Omni lets you edit the video through conversation. Change the camera angle in step 3, and it still