
Pasquale Simonetti
3.3K posts

Pasquale Simonetti
@pasqualesim
IT Developer
Milky Way Katılım Nisan 2012
2.2K Takip Edilen562 Takipçiler

Pasquale Simonetti retweetledi

Ali's Wan 2.2 is absolutely insane 🤯
Map your voice and motion into a video clip featuring anyone with a single photo.
In this video I shape-shift between an avatar woman, Matthew Mcconaughey, Anne Hathaway, Martin Luther King Jr., and Bane.
Tools used:
- Wan 2.2 for video gen
- FreePik for image gen
- ElevenLabs for voice cloning and voice changing
Want the full prompting guide and a step-by-step tutorial video where I walk through exactly how I built it?
Like, RT, and comment "REALITY", and I'll DM it to you.
(must be following so I can dm you)
English
Pasquale Simonetti retweetledi
Pasquale Simonetti retweetledi
Pasquale Simonetti retweetledi
Pasquale Simonetti retweetledi

My brain broke when I read this paper.
A tiny 7 Million parameter model just beat DeepSeek-R1, Gemini 2.5 pro, and o3-mini at reasoning on both ARG-AGI 1 and ARC-AGI 2.
It's called Tiny Recursive Model (TRM) from Samsung.
How can a model 10,000x smaller be smarter?
Here's how it works:
1. Draft an Initial Answer: Unlike an LLM that writes word-by-word, TRM first generates a quick, complete "draft" of the solution. Think of this as its first rough guess.
2. Create a "Scratchpad": It then creates a separate space for its internal thoughts, a latent reasoning "scratchpad." This is where the real magic happens.
3. Intensely Self-Critique: The model enters an intense inner loop. It compares its draft answer to the original problem and refines its reasoning on the scratchpad over and over (6 times in a row), asking itself, "Does my logic hold up? Where are the errors?"
4. Revise the Answer: After this focused "thinking," it uses the improved logic from its scratchpad to create a brand new, much better draft of the final answer.
5. Repeat until Confident: The entire process, draft, think, revise, is repeated up to 16 times. Each cycle pushes the model closer to a correct, logically sound solution.
Why this matters:
Business Leaders: This is what algorithmic advantage looks like. While competitors are paying massive inference costs for brute-force scale, a smarter, more efficient model can deliver superior performance for a tiny fraction of the cost.
Researchers: This is a major validation for neuro-symbolic ideas. The model's ability to recursively "think" before "acting" demonstrates that architecture, not just scale, can be a primary driver of reasoning ability.
Practitioners: SOTA reasoning is no longer gated behind billion-dollar GPU clusters. This paper provides a highly efficient, parameter-light blueprint for building specialized reasoners that can run anywhere.
This isn't just scaling down; it's a completely different, more deliberate way of solving problems.

English

Nano Banana + Veo 3 + N8N is absolutely WILD
This AI system creates unlimited UGC videos with the same character across every single piece of content.
No casting calls. No talent agencies. No $10K video shoots.
Just one product photo turned into an entire video campaign with perfect character consistency.
Perfect for e-commerce brands and agencies who need authentic-looking content at scale without the headaches of managing real creators.
Here's the magic:
→ Submit one product image
→ System analyzes product details and optimal presentation angles
→ Nano Banana generates the same realistic character in multiple scenarios
→ Veo 3 creates video scenes with identical talent across all clips
→ Automation stitches everything into polished campaign-ready content
→ Download complete UGC library in minutes
Same face, different scenarios, every time.
Built 100% in N8N automation.
While competitors spend weeks coordinating with creators and managing inconsistent content, you'll have an entire campaign ready before lunch.
Want the complete system?
Comment "BANANA" + RT + Like
I'll DM you the entire N8N workflow
(Must be following so I can DM)
Skip this and keep juggling 15 different creators for basic consistency.
English

I just trained a custom GPT that turns any idea into a Nano Banana JSON prompt
just describe the picture you want and it will:
– plan the photo like a creative director
– offer pro-level suggestions
– write the entire JSON prompt for you
all you have to do is copy and paste the JSON and you'll get images that look like a creative director made it
comment "nano" and I'll DM it to you (must be following)
English

@vonderleyen “Freedom of corruption” slogan sounds more credible
English

@rauchg European bureaucrats have ruined it all. I’m pro a united Europe but the current EU is shameful and I hope it will cease to exist as soon as possible
English

This is exactly what is happening at MASSIVE SCALE!!
Wall Street Mav@WallStreetMav
We are taxed. Taxes go to USAID. Leftists create a NGO. NGO is funded by USAID. Leftists in NGO do 0 real work. NGO pays Leftists huge salaries. Leftists donate salaries from NGO to Dems. It's why Dems are freaking over Trump & Elon killing USAID. Simple!
English







