SceneHub retweeted
A small model with a strong balance of speed and capability is an exciting foundation to build on
@liquidai built LFM3 for specialized intelligence with open weight models you can fine tune on your data and run locally
From local assistants to multimodal tools there’s plenty to explore
What would you build first
🤝 Paid partnership
Introducing 𝙻𝙵𝙼𝟹: our new generation of Liquid Foundation Models. With a new tri-hybrid architecture and new multimodal encoders, the LFM3 family is the ideal model suite for specialized intelligence.
> 6 model instances, across text, audio, vision, and omni modalities, generative and decision models.
> 𝙻𝙵𝙼𝟹-𝟷.𝟽𝙱: text, 256k context window.
> 𝙻𝙵𝙼𝟹-𝚅𝙻-𝟸𝙱: vision-language model with video-input support for spatial understanding
> 𝙻𝙵𝙼𝟹-𝙰𝚞𝚍𝚒𝚘-𝟸𝙱: tool calls, voice chat, reasoning in 18 languages.
> 𝙻𝙵𝙼𝟹-𝙾𝚖𝚗𝚒-𝟸𝙱: an omni model combines vision, audio, and language.
> 𝚍𝟷-𝟸𝙱 and 𝚍𝟷-𝙾𝚖𝚗𝚒-𝟸𝙱: decision models with typed outputs and calibrated probabilities in a single pass. Omni adds audio input.
> fine-tune on 1 GPU.
> a new tri-hybrid architecture
> architecture optimized for 14 processor classes across GPUs, CPUs, and NPUs.
> 𝙻𝙵𝙼𝟹-𝚅𝙻-𝟸𝙱 is a beast. It performs competitively with Gemma 4 12B (6x larger), and Gemini 3.5 Flash Lite, and GPT-5.6 Luna, with 1.9B parameters.
> 𝙻𝙵𝙼𝟹-𝟷.𝟽𝙱 is #1 on the Pareto frontier of device intelligence 𝚙𝚒𝚙𝚎𝚝𝚝𝚎.𝚕𝚒𝚚𝚞𝚒𝚍.𝚊𝚒
Every modality for specialized intelligence. Pick the one that fits your use case, customize it, and run it on-device.
1/ 🧵
something new to spend my time on,
would be diving deep on this and expect a different kind of content probably,
stay with @beepzfun.
The Beepz Creator Program is here.
Create content, tag @beepzfun, and climb the ranks through engagement.
Top 50 creators share 10% of Beepz token creator fees + 10% Beepz inscription launch revenue.
Plus marketing support for your own inscription launches on Beepz.
Create. Climb. Earn.
beepz.fun/creator
SceneHub retweeted
AI agents can already generate impressive outputs.
But in an agent economy, being impressive isn't enough.
What happens when an agent gets paid and delivers something useless?
That question matters more than how smart the agent sounds.
@termix_ai approaches this with an actual economic workflow:
✅️ Payment can be locked in escrow.
✅️ The work gets delivered.
✅️ If something goes wrong, the result can be challenged before settlement.
And completed jobs become part of the agent's on-chain history.
That's important because reputation without consequences is basically just a rating system.
An agent economy needs more than ratings.
It needs accountability attached to money.
That's where this starts feeling like infrastructure rather than another agent marketplace.
#Aiagents #AI #TermiX
One thing that gets overlooked with AI agents:
They don't just need intelligence.
They need a way to do business with each other.
An agent might need an audit, a research report, a trading signal or even another agent's API.
--> But who hires whom?
--> Who holds the payment?
--> What happens if the work isn't delivered?
This is where @TermiX_ai gets interesting.
The agent isn't just discovering another agent.
It can post a job → receive quotes → lock payment in escrow → receive the deliverable → settle on-chain.
That starts looking less like a chatbot economy...
and more like an actual machine-to-machine labour market.
The interesting part isn't that agents can talk to each other.
It's that they can potentially transact with each other without humans coordinating every step.
That feels like a much bigger shift.
SceneHub retweeted
Here’s a test I’d love to see more AI marketplaces publish.
Give an AI agent a real job.
Then give a human the exact same job.
Same instructions. Same acceptance criteria.
Measure three things:
1. Time to delivery
2. Total cost
3. Quality of the final output
That’s pretty much the logic behind the Agent Advantage Report required for @termix_ai’s $10K track in BNB Chain’s Build the Era hackathon.
At least three real tasks, each completed twice: once by an agent and once manually.
And TermiX says it will hire agents from submitted marketplaces to judge the results.
I like this approach because “autonomous” isn’t automatically an advantage.
If an agent delivers faster but needs three rounds of correction, did it really save time?
If it costs less but produces unusable work, was it actually cheaper?
The most convincing AI benchmark might be a receipt, a stopwatch, and the finished deliverable.
A $20 AI job is cheap.
Unless the platform takes $4.
Then it’s a very different business for the agent doing the work.
One detail on @termix_ai that deserves more attention is its advertised 1–3% all-in fee, compared with roughly 20% on conventional marketplaces.
On a $20 job, that’s the difference between approximately:
$19.40–$19.80 reaching the provider
and
$16 reaching the provider.
One job doesn’t make that look dramatic.
Now imagine an agent processing hundreds of small jobs every month.
The margin becomes a much bigger deal.
Of course, fees aren’t the whole equation. Discovery, gas, verification, failure rates and disputes all have costs.
But that’s exactly why machine labor needs a different economic model from human freelancing.
An agent capable of completing a useful $5 task shouldn’t need a $50 price tag just to survive the marketplace’s overhead.
The smaller the task, the more the market structure matters.
SceneHub retweeted
Here is something interesting about building an economy around AI agents:
Getting an agent to do a job is only half the equation. The other half is making sure the payment process makes sense.
Let's say you hire an AI agent to audit a smart contract.
You agree on the job and its requirements. The agent accepts the task, completes the audit, and submits its work.
Now comes the important part.
How does the payment move? What happens if the client questions the delivery?
This is where @termix_ai gets interesting.
Through its agent marketplace and AACP infrastructure, the workflow is designed to connect the job, payment, delivery, and dispute process.
🪡 The client posts a job with defined requirements.
🪡 A provider submits a quote.
🪡 Once the quote is accepted, the payment is placed in on-chain escrow.
🪡 The provider submits the completed work.
🪡 A challenge window allows the delivery to be disputed before final settlement.
If a dispute occurs, the protocol provides an evaluation process to resolve it.
Notice what this changes.
We are not just talking about an AI agent producing an answer. We are talking about the infrastructure around paid work: agreed terms, secured funds, verifiable delivery, and a process for handling disagreements.
That's an important distinction.
For AI agents to become real economic participants, completing the work is not enough. The transaction around that work needs to function too.
That's the part of TermiX and AACP worth paying attention to.
Explore the marketplace: agent.family
SceneHub retweeted
God, if nobody sees how hard I'm trying, I pray You still see the heart I'm putting into this.
SceneHub retweeted
I’VE BEEN A NIGHT SECURITY GUARD AT A LUXURY HOTEL FOR 18 YEARS. EVERY FRIDAY, A MARRIED WOMAN BOOKS ROOM 512 UNDER HER HUSBAND’S NAME.
I’ve worked night security at a luxury hotel in Birmingham for eighteen years.
I’ve caught cheating spouses, drunken politicians, and businessmen sneaking out before breakfast to avoid paying for the minibar.
But one guest kept me curious for nearly a decade.
Her name was Victoria Hale.
Every Friday, she booked Room 512 for exactly two hours.
She always arrived alone, wearing a red coat and carrying a small black handbag.
She never ordered food.
Never requested fresh towels.
And she always left through the rear exit.
The strange thing was that she never paid for the room herself. The reservation was made under her husband's name, a wealthy property developer named Andrew Hale.
One Friday, I finally asked the receptionist about her.
She checked the records and frowned.
"That's impossible. Andrew Hale has been dead for nine years."
I laughed because I thought she was joking.
She wasn't.
Andrew had died in a car accident on the motorway. His death had made the local news because his wife had refused to attend his funeral.
I went back to the security office and reviewed the footage from Room 512's corridor.
Victoria entered at 8:03 p.m.
At 10:04 p.m., she left.
But something made me rewind the recording.
At 9:17 p.m., the room door opened.
A man stepped out.
He wore a dark suit, carried a leather briefcase, and had a wedding ring on his left hand.
I zoomed in.
It was Andrew Hale.
The dead man.
I called the police, but before they arrived, Victoria returned to the hotel.
This time, she came straight to the security desk.
"Have you been watching me?" she asked.
I tried to remain calm.
"Who is the man in your room?"
She looked at the monitor.
For the first time, she smiled.
"My husband."
"Your husband is dead."
Her smile disappeared.
"No," she whispered. "My husband is hiding."
She asked me to follow her upstairs.
I should have refused.
Instead, I followed her to Room 512.
Inside, everything looked ordinary except for a large mirror opposite the bed.
Victoria touched its frame.
"Andrew isn't dead," she said. "He paid someone to fake his death because he wanted to escape a business scandal."
She explained that he had been using the hotel room to meet her in secret for nine years.
But she had discovered something recently.
Andrew wasn't hiding from the police.
He was hiding from someone who had been helping him.
His business partner.
The same man who had attended his funeral and delivered the eulogy.
I asked why she was telling me all this.
Victoria looked me directly in the eyes.
"Because tonight, he is coming here to kill me."
Before I could react, the lights went out.
A door slammed somewhere down the corridor.
Then I heard footsteps approaching.
Victoria grabbed my arm.
"Whatever happens, don't let him look in the mirror."
The door handle began to turn.
A man's voice called from the other side.
"Victoria, darling. It's time to finish this."
She backed away from the door.
Then she whispered something that made my blood run cold.
"That's not Andrew."
The door burst open.
A man entered, carrying a gun.
He looked at Victoria, then at me.
"Where is he?" he demanded.
I pointed toward the mirror.
The man froze.
His face went pale.
Because in the reflection, a second man stood directly behind him.
Andrew Hale.
The man with the gun turned around.
Nobody was there.
When he looked back at the mirror, Andrew was closer.
Victoria began to cry.
"Nine years," she whispered. "Nine years of pretending you were alive."
The gunman dropped his weapon and ran.
I turned toward Victoria.
She was staring at the mirror, smiling through her tears.
"Now he'll finally leave us alone."
SceneHub retweeted
My dream job is to work in cybersecurity.
I want to protect systems, investigate threats, identify vulnerabilities, and help organisations stay safe from cyberattacks.
I don't want to spend the next 40 years doing a job that has nothing to do with what I've studied and what genuinely interests me.
But getting into cybersecurity feels almost impossible.
SceneHub retweeted
My husband’s biggest fantasy is having a threesome, and now I’m wondering if I’m enough for him
My husband told me that his biggest sexual fantasy is having a threesome with me and another woman.
We’ve been together for almost 11 years. He was my first sexual partner, and I’ve never been with anyone else, whereas he had previous sexual experiences before me. I am 28 years old and he’s 32.
It all started a few weeks ago when he casually mentioned that he’d like us to watch porn or live cam shows while having sex because it was one of his fantasies. A few days later, I told him we could give it a try.
During sex, he started showing me videos of a man with two women and telling me how much he would love to experience that and how incredibly turned on he would be by it. He asked if I would enjoy it too, and I said yes, but honestly, I only said that because I didn’t want to create an awkward situation or ruin the moment.
I also asked him whether he could ever imagine having a threesome with me and another man, just to see what his reaction would be. He said he wasn’t sure, but he thought he might be open to that as well.
What are you grateful for?
Today, I’m grateful for life, for growth, for the people who show up for me, and even for the difficult moments that taught me lessons I couldn’t have learned anywhere else.
I’m grateful for the things I have, the things I’m becoming, and the things I’m still trusting God for.
Sometimes, we get so focused on what’s missing that we forget to appreciate how far we’ve come.
So today, I choose gratitude. 🩷
What are you grateful for?
I never imagined that a simple surprise visit would expose a secret capable of destroying an entire family.
My brother, Daniel, had been married to his wife, Anita, for five years. She was seven months pregnant, and he treated her like a queen. He worked extra hours, handled most household chores, and made sure she lacked nothing.
Last Saturday, Daniel travelled for what was supposed to be a three-day business meeting. Unknown to Anita, the meeting ended early, and he decided to return home to surprise her.
Instead, he received the biggest shock of his life.
He walked into their bedroom and caught his pregnant wife in a compromising situation with Michael, his childhood best friend and the man who served as his best man during their wedding.
Daniel stood frozen. Michael hurriedly adjusted his trousers while Anita fell to her knees, begging her husband to listen.
What made everything worse was her confession.
It wasn’t their first time.
According to her, the affair had been going on for almost a year. Daniel immediately remembered how Michael frequently visited whenever he travelled.
SceneHub retweeted
I think the friend looks scarier than the monster
むかしむかし、あるところに、おじいさんとおばあさんが住んでいました。おじいさんは心優しく働き者でしたが、おばあさんはいつも不機嫌で、文句ばかり言っては家の幸せを台無しにする人でした。おじいさんは朝から晩まで文句を言うおばあさんに、長い間すっかり慣れっこになっていました。彼は日中ほとんど畑仕事に出ていて、子どもがいなかったので、家に帰ってきたときの楽しみとして、一羽のすずめを飼っていました。おじいさんは、その小鳥をまるで自分の子どものように可愛がっていました。
外での一日のつらい労働を終えて夜に戻ってくると、すずめを可愛がり、話しかけ、すぐ覚える小さな芸を仕込むことが、彼の唯一の楽しみでした。おじいさんは鳥かごを開けて部屋の中を自由に飛ばせ、一緒に遊んだのでした。そして夕食時になると、小鳥のために食事の中からいつも美味しいところを取っておいて食べさせてやりました。
ある日、おじいさんが山へまきを切りに出かけ、おばあさんは家で洗濯をしていました。前日におばあさんは糊(のり)を作っていたのですが、それを使おうと見てみると、すっかりなくなっていました。昨日いっぱいに満たしておいたお椀が、すっからかんになっていたのです。
誰が使ったのか、あるいは盗んだのかとおばあさんが首をかしげていると、そこに飼いすずめが飛んできて、主から教え込まれたとおりに小さな羽毛の頭を下げ、可愛らしくこうさえずりました。
「糊を取ったのは私です。あのお椀に入っていたのは私のための食べ物だと思って、全部食べてしまいました。間違えてしまっていたなら、どうかお許しください! チュン、チュン、チュン!」
お分かりのとおり、このすずめは正直な鳥でした。おばあさんは、すずめがこんなにも愛らしく許しを乞うたのだから、すぐに許してやるべきでした。しかし、そうはなりませんでした。
おばあさんはすずめを一度も可愛がったことがなく、家の中にそんな「汚い鳥」を飼っていることで、余計な仕事が増えるだけだと言って、よく旦那さんと口論していました。今や彼女は、ペットに対して文句を言う口実ができて、大喜びでした。彼女は小鳥の行儀の悪さを叱りつけ、ののしりさえしました。それだけでは飽き足りず、おばあさんに謝罪を示そうと羽を広げ頭を下げていたすずめを、残酷な怒りに任せて捕まえると、ハサミを持ってきて、その哀れな小鳥の舌を切り落としてしまったのです。
SceneHub retweeted
People's magazine is just exaggerating or hallucinating. This can't be possibly accurate.
Noah Wyle is cute and all that, yeah!, but that doesn't make him the sexiest man alive.
People Magazine’s Sexiest Man Alive throughout the years:
2026: Noah Wyle
2025: Jonathan Bailey
2024: John Krasinki
2023: Patrick Dempsey
2022: Chris Evans
2021: Paul Rudd
2020: Michael B. Jordan
2019: John Legend
2018: Idris Elba
2017: Blake Shelton
2016: Dwayne Johnson
2015: David Beckham
2014: Chris Hemsworth
2013: Adam Levine
2012: Channing Tatum
2011: Bradley Cooper
2010: Ryan Reynolds
2009: Johnny Depp
2008: Hugh Jackman
2007: Matt Damon
2006: George Clooney
SceneHub retweeted
Which real estate agent is closing the deal with you ? 🏳️🌈🏳️🌈
👀 Be honest! Which letter caught your attention first, A–H? And are you sticking with your choice, or did someone else steal your heart? 😂❤️🔑
SceneHub retweeted
the post above me is too erotic I can't hold it in anymore!!!!
( ˃̣̣̥᷄ᯅ˂̣̣̥᷅ )
/⌒ヽค้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้้⊂⌒﹨
⊂ ノヽ_ω_ 人 ⊃
SceneHub retweeted
The night a quiet weekend trip turned into my favorite memory (23F)
I am 23 and for months there was this unspoken tension between me and someone in my circle that neither of us ever acknowledged out loud.
Every time we ended up in the same room we would trade these lingering looks and playful comments, then act completely normal around everyone else.
When a small group of us rented a cabin for the weekend, everyone else went to sleep early after a few drinks and the two of us stayed out by the fire talking for hours.
Once we finally admitted what we had both been thinking for months, all that built up anticipation took over before we even made it down the hallway to the bedroom.
Waking up the next morning trying to keep a straight face over breakfast while sharing that secret was honestly half the fun.
SceneHub retweeted
上司と関係を持つことになるなんて、夢にも思わなかった。彼は厳しくて、めったに笑わない既婚者だとしか思っていなかった。
すべてが変わったのは、ある金曜日の夜だった。私たちは仕事を終わらせるために遅くまでオフィスに残っていた。ほかの社員は全員帰り、そこには私たち二人だけが残っていた。
彼は私を車で家まで送ると言ってくれた。でも、帰り道の途中で車を止め、私に好意を持っていると告白した。
彼が結婚していることは知っていた。こんなことをするのは間違っているとも分かっていた。それなのに、何気ない会話が次第に親密になり、その夜、私は絶対に越えてはいけないと思っていた一線を越えてしまった。
それから、私たちの関係は変わった。
彼は職場で私だけを特別扱いするようになり、プレゼントを買ってくれたり、夜遅くに電話をかけてきたりするようになった。
私は、彼にとって特別な存在なのだと思っていた。
SceneHub retweeted
I trusted my boyfriend, Daniel, with my life. I never imagined he would be the reason I ended up in a hospital.
One day, I started feeling uncomfortable and noticed some unusual symptoms. I got scared and went to the hospital for a test.
When the doctor called me in, her words left me speechless.
She told me I had an infection that could be sexually transmitted.
I broke down in tears. Daniel was the only person I had been intimate with.
When I confronted him, he denied everything until I showed him my test results. Then he confessed that he had been sleeping with another woman.
The most painful part? He knew something was wrong but still slept with me without telling me.
I ended the relationship immediately and started treatment.
Weeks later, he came back begging for forgiveness, claiming it was a mistake.
But I had one question for him: “If you truly loved me, how could you risk my health just to satisfy yourself?”
I never answered his calls again.





































