←
Leon Lin19,797 posts
Leon Lin avatar

Leon Lin

@LexnLin

highschool | creative ai github.com/Leonxlnx/

◷ Joined March 2025▧ 2,586 Photos and videos
Leon Lin avatar
Leon Lin@LexnLin·Oct 13

We rebuilt TasteSkill from scratch And today we are introducing TasteSkill v2 v2 gives your coding agent an art direction, assets and a browser check to verify its work npx skills add Leonxlnx/taste-skill check out more examples below!

1:04
Leon Lin avatar
Leon Lin@LexnLin·2h

this wavy filter over image generation so it looks like it made it real

0:08
Leon Lin avatar
Leon Lin@LexnLin·2h

using my reference image as the build and zooming around and adding random filters

0:29
Leon Lin avatar
Leon Lin@LexnLin·2h

Didn't want to judge GPT Astra 6.2 and Sol 6.2 too early but now I will after testing them: Both Astra 6.2 and Sol 6.2 are lazy models. They do not follow instructions properly and take the easiest route (against instructions) that no other model would take. I've been building 3D things with Astra 6, Fable 5.2 and Opus 5.5. They ALL created the result either using Blender or entirely using three.js. But the GPT 6.2 models SOMEHOW, even though I CLEARLY said "USE ONLY THREEJS and ONLY CODE no assets", fucking took my reference images and put a wavy fisheye effect on the part of the image where the water is to make it look like real water. And then zoom and pan around in the images as if it was showing what it really built. In one aspect it is NOT lazy: it puts random asf slop UI around the 3D thing. I told it to render a video and give that video to me in the end. Only gave me the animated images stitched together playable on a website, not a video. The model also LOVES to cheat when building 3D demos with image generation. It puts them everywhere. I'd stick to Claude for now. This shit's not usable. Maybe for some more engineering focused users cuz I haven't tested that area yet but for my work which is mostly involving UI, 3D and taste this is NOT IT. Maybe computer use is cool with it but for me it is unusable. Wanna repeat for ME it's not good. You might have better experiences. I'll run some tests again really trying to avoid letting it cheat and post the results later if I get any good results that are worth posting. (below are some of the cheating outputs)

running some tasks overnight with Astra 6.2 and Sol 6.2 and will report back how strong they are in 3D demos/design taste

Post attachment
Leon Lin avatar
Leon Lin@LexnLin·10h

I love the details Fable 5.2 adds to its paintings

Post attachmentPost attachmentPost attachmentPost attachment
Leon Lin avatar
Leon Lin@LexnLin·13h

running some tasks overnight with Astra 6.2 and Sol 6.2 and will report back how strong they are in 3D demos/design taste

Post attachment
Leon Lin avatar
Leon Lin@LexnLin·16h

no pre-made 3D assets were used. only thing that is not code is the sound/bg music: "Passage Of Time" by Scott Buckley (CC BY 4.0), sfx from freesound repo is here: github.com/Leonxlnx/mechanic…

Leon Lin avatar
Leon Lin@LexnLin·16h

A mechanical moon in Blender by Fable 5.2

0:45
Leon Lin avatar
Leon Lin@LexnLin·19h

I don't get it. Why is OpenAI showing that GPT 6 Astra was SO bad in comparison to 6.2 Astra?? (i mean just look at the quoted 6 Astra example) Also I believe that 6.2 Astra can do WAY better as well.

Post attachment

GPT-6 Astra recreated the Palace of Fine arts in Blender. This is favorite building in San Francisco because it was built for the World's Fair in 1914 where the steam locomotive and telephone were showed off. It was a time when technology gave us all a deep sense of optimism for the future. I feel like some of that sense has been lost since, but I hope models like astra can help restore it, and push us towards a more hopeful future.

0:30
Leon Lin avatar
Leon Lin@LexnLin·21h

ChatGPT Pro 200 subscribers can test ultrafast with GPT 6.2 Sol now as well!

Starting today, GPT-6.2 Astra and Sol are available in ChatGPT Work and Codex for Plus, Pro, Enterprise, and Edu users. Pro 500 subscribers can use Ultrafast with either model. For the first time, we’re also bringing Ultrafast to Pro 200 subscribers, starting with GPT-6.2 Sol. Developers can access gpt-6.2-astra and gpt-6.2-sol in the OpenAI API, with Ultrafast available.

Leon Lin avatar
Leon Lin@LexnLin·Nov 17

a drop of ink turns into a dragon, made with fable 5.2

0:45
Leon Lin avatar
Leon Lin@LexnLin·Nov 17

Makes me really sad to see so many bot replies on my posts :/ Anything I can do against that? When will this issue finally be fixed?

Post attachmentPost attachmentPost attachmentPost attachment

Little experiment with Fable 5.2 and Opus 5.5: Creating a working ADE for Claude Code and Codex with good UI/UX in 3 prompts This is the result! github.com/Leonxlnx/sable

0:46
Leon Lin avatar
Leon Lin@LexnLin·Nov 17

Little experiment with Fable 5.2 and Opus 5.5: Creating a working ADE for Claude Code and Codex with good UI/UX in 3 prompts This is the result! github.com/Leonxlnx/sable

0:46