Pinned Tweet
Just got 44K+ monthly views on YouTube with 700+ subs
About to hit 1K soon.
DMs are open if you want to sponsor a video 😉
7
804
Just tested the Qwen 4 Max preview and had it build a 3D destruction playground using Three.js.
It took around 45-60 minutes, but the overall quality is actually pretty nice.
Performance wise, it feels quite similar to Qwen 3.8 Max.
1
121
just tested MiniMax H3 Image on a free T4 at 480p Quality
prompt is about underwater office
took about 8m and 11s
pruned variant still looks pretty good
worth a try on Kaggle
1
4
133
Just tested Gemini 4 Carbon in AI Studio and it built this Black Hole game from a single prompt.
Nice attention to detail and model feels token efficient for coding.
2
78
Tested Muse Ember 1.0 on XHigh reasoning.
Built this 3D jungle boat game in an hour.
Used 31% of my 5 hour limit and 10% of my weekly limit on the $5 plan.
Might be a good companion to models like Opus and Fable.
4
83
Meta just released Muse Ember 1.0.
Beats Claude Haiku 5.5 and GPT 6 Luna on several benchmarks.
And it's priced at just $0.10/M input tokens and $0.50/M output tokens.
today we released our smallest model yet: Muse Ember 🔥
Muse Ember offers nearly Muse Spark 1.3 (max) intelligence at a 17x cheaper price point (per AA)
Muse Ember is the first model we’ve released trained using Muse Wonder Full
excited for y’all to try on API and muse code!
91
Jigs retweeted
⚡️ 免费 2×T4 也能跑剪枝版 Qwen3.8 Flash Next Coder
Kaggle + Strata，生成 SVG 约 28 tok/s。
本地还没装机前，可以先拿免费卡摸一下手感。
温馨提示：单次演示，换机别当保底。
来源：@debugsenpai👇
Just ran Qwen 3.8 Flash Next Coder (pruned 125B model) on kaggle's free 2x T4 with Strata.
Generated this SVG at 28 tok/s.
102
Just ran DeepSeek V4.1 Flash 1017 on an L4 GPU using Strata.
Generated a complete SVG Rocket illustration at 14.33 tokens/sec.
Running a 734B parameter model on just 24 GB GPU is crazy.
6
98
Closed models are fine.
But Qwen 4 with open weights?
That’s what we’re all waiting for.
Qwhen? Qsoon!
For our friends outside China, the Chinese text in the poster comes from Zhuangzi, an ancient Chinese classic:
"Far in the northern sea lives a fish named Kun (pronounced like Qwen). So immense is its size that no one knows how many thousands of miles it spans."
113
Just tested Ling 3.2 Flash.
It one shotted this voxel pagoda scene, and the quality is insane 💀
Honestly, it looks even better than Tencent HY4.
2
3
82
Just tested Ling 3.2 Flash for free through their chat platform.
Built a third person shooting game with it. Took a few extra prompts to improve the result, but overall, pretty decent for a 560B parameter model.
Is Ling 3.2 Flash better than Tencent HY4? 🤔
1
2
77
Just tested Speridlabs Idol, an image to 3D model
Converted an AI generated treasure chest image into a 3D model.
Ran it on a free Kaggle T4. Want a tutorial to run it?
1
100
Just tested Gemini Omni Flash 1.2 through the API.
Generated a 5 second cinematic car video in 25.5 seconds.
Should I compare it with MiniMax H3?
🤖 Made with AI
1
3
183
Just tested Tencent HY4 and used it to create a 3D voxel art scene.
The model is definitely better than the preview version.
Is HY4 really GPT 6.1 Sol level?
1
3
163