Noir Abbé

@OpenGradient $OPG #OPG
I bought OpenGradient tokens at 2am during a market pump and didn't actually know what I was doing for three weeks. Just threw money in because the ecosystem narrative felt right, then went back to sleep.

The embarrassing part? I'd check the price obsessively but never once looked at what was actually happening on-chain. My OpenGradient bag was just sitting there. No staking. No governance votes. Nothing. I was basically paying rent on capital that wasn't working for me.

Last month I had coffee with someone who's actually running a node in the OpenGradient network, and he asked me a simple question: "Are you participating or just spectating?" Hurt a little, not gonna lie. Because the answer was obvious. I was speculating on other people's work.

That conversation forced me to actually open the governance dashboard. Saw settlement happening. Real AI requests being processed. People making decisions about network upgrades. Suddenly my OpenGradient tokens weren't just numbers on a screen they represented actual voting power, actual network stake, actual skin in the game.

I started small. Staked some OPG. Paid attention to one governance vote. Felt weird at first. But it changed how I think about my position completely. I went from passive bag holder to someone actually invested in whether OpenGradient succeeds.

Most people's crypto holdings are just sleeping assets they hope go up. Are yours actually doing anything?

Real talk: Your OpenGradient bags are only worth what the network actually does. So what's your role?

$VELVET $PIEVERSE

Are you creating utility in OpenGradient or just watching it grow?

#OPG $OPG
Last week I had to retry a payment after an AI inference had already finished. The computation was complete, but the payment settlement failed on the first attempt. It was a small issue, yet it reminded me that delivering inference isn't the same as completing the economic transaction. Until payment settles, the workflow is still incomplete.

My first thought was that better regulation would solve this kind of problem indirectly. If MiCAR gives projects more regulatory clarity and broader market access, I assumed that would naturally strengthen token demand. After thinking through the payment flow again, I realized I was mixing two very different things.

What actually matters is whether people keep using the protocol. With @OpenGradient , demand grows when applications repeatedly request inference, users successfully pay for those services with OPG, and validators or stakers continue securing the network. Regulatory classification can reduce uncertainty and make participation easier, but it doesn't create economic activity by itself. The token only becomes more valuable if the infrastructure keeps processing real workloads and payments continue settling reliably. 🚀
That changed how I evaluate projects. I spend less time watching trading volume or debating classifications and more time looking at recurring inference requests, successful payment settlements, and sustained protocol participation because those are the signals that reflect actual usage.

Which on-chain or protocol metrics do you think best capture real token demand beyond price and trading volume?

$PIVX $NFP

📊 Best demand signal?

#OPG
Many AI projects introduce a token first and search for utility later. @OpenGradient takes a different direction by placing $OPG at the center of how the network functions, making its value increasingly tied to AI activity rather than narrative alone.

The core idea is simple: if the protocol succeeds in attracting meaningful AI workloads, OPG becomes the medium through which that demand is expressed across the network.

At the protocol level, OPG facilitates payments for AI services while staking helps secure participation and align incentives between users and operators. From an economic perspective, growing AI usage can translate into recurring token demand instead of relying solely on speculative trading. Governance adds another layer by allowing participants to influence protocol upgrades and long-term network priorities.

None of this guarantees success. AI infrastructure is becoming increasingly competitive, and strong token design cannot compensate for weak adoption or limited developer traction.

That said, OpenGradient is building around a principle that is becoming harder to ignore: tokens tend to hold stronger long-term relevance when they sit inside the protocol's economic flow rather than outside it.

The real question isn't whether OPG has utility today, but whether OpenGradient can create enough AI activity for that utility to compound over time. 🤔

$SYN $PORTAL

What do you think is the biggest driver of long-term token value for AI protocols like OpenGradient?

$OPG #OPG
For a long time, I looked at crypto roadmaps the same way most people do: more releases meant more progress.

But @OpenGradient makes me think the better question is whether each new layer actually strengthens the one next to it.

There are already thousands of models available on the network and millions of recorded inferences. Those numbers show that something is happening. At the same time, activity alone does not automatically translate into lasting demand. Especially while the network is still developing, it is difficult to know how much of that usage will remain over time.🔍

What I find interesting is the relationship between the different parts of the ecosystem. Models need compute to be useful. Compute needs verification to be trusted. Verification needs payments to create incentives. And payments only become meaningful when applications provide enough value for users to return.

That creates a simple but important reality: adding more components does not necessarily make the network stronger. If one part of the chain struggles, the rest can appear healthier than they really are.

This is why I think OpenGradient’s roadmap is less about shipping features and more about building connections between them. The real milestone is not another launch or integration. It is reaching a point where usage in one area naturally drives usage in the others.
When that happens, OPG demand stops depending on expectations and starts reflecting actual network activity.

So the question is not how many pieces OpenGradient can add. It is whether those pieces can eventually create a loop that keeps users, developers, and demand moving in the same direction.

What is the biggest driver of sustainable OPG demand?

#OPG
I used to think the biggest challenge for decentralized AI was getting models to run outside centralized infrastructure.

Now I’m not so sure.

For @OpenGradient , the more interesting question may be whether verification can stay affordable as AI models become larger and more complex.

The reason is straightforward: as model complexity increases, the amount of work required to prove and verify outcomes grows too. Decentralized AI doesn't just need computation it needs a reliable way to validate that computation without trust. That creates a different kind of scaling challenge. If verification becomes increasingly expensive, participation can start to narrow toward the actors with the resources to absorb those costs. In that scenario, decentralization remains technically possible but becomes harder to sustain economically.

What makes this especially relevant for $OPG is that its value proposition sits at the intersection of security, transparency, and decentralized coordination. Those benefits only hold if verification remains efficient enough to support growth without introducing excessive overhead. A fair counterpoint is that proving systems, specialized hardware, and verification techniques are still evolving rapidly. If those technologies improve faster than model complexity grows, today's concerns may look overstated.

To me, the long-term question isn't whether decentralized AI can work. It's whether the cost of verifying intelligence can stay low enough to keep decentralization practical.

As decentralized AI networks mature, which will improve faster: verification efficiency or model complexity?
$ATM $HEI

#OPG $OPG
I’ve watched enough “AI shipped to prod” moments to recognize the smell🔥

It’s the same pattern every time:

➤ New model drops.
➤ Metrics look good.
➤ Everyone celebrates.
➤ Two weeks later someone asks, “why did it do *that*?”

And suddenly nobody can replay the exact conditions.

↪ Different weights.
↪ Different prompt template
↪ Different retrieval index.
↪ Even the “memory” changed.

In crypto, we call that a missing audit trail. In AI, we call it “iteration.”

Quiet detail most people ignore: **most AI systems can’t produce a clean receipt for a single answer**. Not just the output of what it saw, which tools it hit, what state it relied on, what version was live, what context got silently swapped.

★ That’s where @OpenGradient has been interesting to track.
★ Not because it claims smarter models.
★ Because it treats inference like something closer to a transaction: anchored to a verifiable state, tied to history, meant to be checked later without hand waving.

The tradeoff is real. Proof isn’t free. Persistence isn’t free. You’re paying for continuity.

But the moment AI touches money or compliance, “cheap answers” stop being the goal.

The real premium is paid when you need to show your work.

✦ It's true ✦

$DEXE $MBL

I Built Stuff on @OpenGradient and Now I Can't Unsee the Problem They Actually Solved

Look, I've been watching the AI infrastructure space long enough to be tired of the pitch. "Decentralized GPU network." "Verifiable inference." It all sounds like the same scaling theater we've heard for three years.

Then I actually tried to deploy an agent that needed to move money based on AI reasoning, and everything changed.

The honest problem I was facing: how do I prove to a smart contract that an AI didn't just make something up? How do I let my agent handle private data without some cloud provider logging every thought it has? How do I sleep at night knowing my customer's financial data is sitting on some server I can't audit?

OpenGradient stopped making me choose between those things.
Here's my actual take: most platforms encrypt some stuff or verify some stuff. OpenGradient went nuts with paranoia in the best way. Your message gets locked down before it leaves your computer. It gets routed through systems specifically designed so no single operator not even OpenGradient can connect your identity to your query. Then the computation happens in an isolated enclave. The proof comes back mathematically verifiable.

I can't trick it into lying. Neither can they.

Is it perfect? No. Does it require you to understand what TEE attestations actually mean? Yeah. But watching the market finally build infrastructure around "verify, don't trust" that's the thing I didn't know I was waiting for.

Most networks ask you to believe. This one lets you check the math yourself. That alone changes everything about how I think about deploying autonomous systems.

#OPG $OPG

The question stopped being "which model wins" sometime last year.

What replaced it was quieter. Harder to spot.

I was watching a DeFi protocol launch an AI agent last month. Cool idea. But I kept thinking who's verifying what the model actually does? Nobody had an answer. Just vibes and "we trust the provider."
That stuck with me.

Here's the thing nobody says out loud: AI output is invisible to most systems. You send a prompt, you get a response. What happens in the middle? You don't know. Could be the model they promised. Could be something cheaper. Could be nothing like what you asked for.

@OpenGradient caught my attention because they're solving this differently. Their inference network puts cryptographic proofs on every call proof of which model ran, proof of what it processed, proof of what it returned. On-chain. Verifiable.

And they've quietly built a hub with over 2,000 open-source models where you can actually pull verified AI compute. No hand-waving.

The founder, Matthew Wang, came from Two Sigma. That background shows they're thinking about this like infrastructure, not hype.

The real shift isn't who builds the smartest model. It's who makes the model accountable.

That's where this is heading.

$OPG
#OPG

Everyone’s impressed by how fast AI agents are getting now.

You see it everywhere tasks done in seconds, workflows collapsing into a single prompt, systems that feel like they’re finally “working for you” instead of the other way around.

It’s easy to focus on that speed.

But something quieter is happening underneath it.

Nobody really stops to ask where all that interaction goes after the screen goes blank. Who stores it. Who can reconstruct it. Who starts to understand you from it over time.

Because AI agents are no longer just tools that execute commands.

They’re starting to become memory systems.

And memory is where things shift.

Not just what you asked once but how you ask, what you avoid, what you repeat, how your decisions evolve. That kind of context builds a version of you that lives outside your control.

@OpenGradient sits in that uncomfortable space where this becomes obvious. The interesting part isn’t the “AI performance” angle it’s the attempt to rethink what happens to context once it exists at all.

Execution, storage, and identity stop being one bundled thing.

They start getting pulled apart.

Technologies like TEEs, OHTTP, decentralized storage, and verifiable compute show up here not as buzzwords, but as weak attempts to answer a simple question: who actually holds the memory when systems start acting on your behalf?

But none of this comes clean.

More privacy usually slows everything down.

More verification adds friction.

More decentralization makes coordination messy.

So the system gets heavier just as it gets more powerful.

And maybe that’s the quiet trade we don’t talk about enough.

We’re not just building faster agents.

We’re slowly building systems that remember us in ways we can’t fully see or easily leave behind.

#OPG $OPG

The longer I watch AI develop, the more I notice a quiet contradiction.

People are trusting AI with bigger decisions every day, yet most have no visibility into what happens between the prompt and the answer.

That’s been one of the industry's blind spots.

Everyone talks about model intelligence, speed, and capabilities. Far fewer conversations focus on verification.

If an AI generates something important, how do you know where it came from?

What data influenced it?

What actually happened behind the scenes?

Crypto users have spent years getting used to a different mindset: don’t just trust the outcome, verify the process.

That’s why @OpenGradient stands out to me.

What’s interesting isn’t simply that it combines AI with decentralized infrastructure. The overlooked detail is its focus on making AI execution observable. Instead of treating the model as a sealed box, the goal is to create systems where the path to an output can be inspected and verified.

After spending time around both crypto and AI, that feels like a more important shift than another benchmark improvement.

The next wave of AI may not be won by the model with the most impressive demo.

It could be won by the systems that make trust measurable.

Because eventually, people won’t just ask whether an AI is right.
They’ll ask if it can show how it got there.
#OPG $OPG

🧠 Most people look at @OpenGradient and immediately see the obvious angle

"AI + blockchain."

I've been following it a bit more closely, and I think that's actually the least interesting part.

The real bet OpenGradient is making is that AI shouldn't be a black box.

Right now, every AI product asks for trust.

Trust the model.

Trust the output.

Trust the company running it.

And for casual use, that's fine.

But the moment AI starts handling value executing trades, managing agents, making decisions, coordinating actions—that trust starts getting expensive.

That's where OpenGradient feels different.

It isn't trying to compete on who has the biggest model.

It isn't trying to convince you that every computation belongs on-chain.

The focus seems to be on something much quieter

📜 proving that AI computation happened as claimed.

That detail gets overlooked because it's not as exciting as model launches or benchmark scores.

But it's arguably the harder problem.

Anyone can show you an AI result.

Showing that the result came from the exact model, exact computation, and exact process being claimed?
That's a different challenge.

👀 The more I think about it, the more OpenGradient feels less like an AI project and more like a trust infrastructure project.

Crypto spent years solving a simple question

"How can strangers agree on what happened?"

OpenGradient is applying thatsame thinking to AI.

Not by replacing AI.

Not by forcing everything on-chain.

But by making important AI actions verifiable instead of simply believable.

Most discussions around AI are still obsessed with intelligence.

OpenGradient seems more interested in accountability.

And if AI agents become as common as many people expect, that distinction might end up mattering more than people realize today.

#OPG $OPG

$SYN $AT

I used to think RAG was just a clean tricknask a question, fetch a few chunks, let the model stitch an answer together.

But when you actually sit with it long enough, you start noticing something uncomfortable: the answer is not the interesting part.

The selection is.👀

In systems like @OpenGradient OpenGradient, what matters isn’t only what the model can say.

It’s what the network lets it see before it speaks.

That part feels almost invisible when everything is centralized. You don’t question the index.

You just assume it’s “the knowledge.”

But knowledge isn’t neutral once it becomes a system.

Retrieval Augmented Generation sounds like grounding models in reality.

In practice, it’s more like routing reality through layers of preference ranking, indexing, caching, latency tradeoffs, node behavior.

And once retrieval is decentralized, something subtle changes. You stop having a single memory.

You get fragments of memory that disagree slightly depending on where you ask from, when you ask, and what the network prioritizes at that moment.

That’s the part most people miss.

Censorship in these systems doesn’t feel like deletion. It feels like something didn’t show up first. Then second.

Then eventually not at all because most users never scroll that far.

And it’s strange how quickly “truth” starts blending into “what arrived fastest.”

The harder question isn’t whether distributed retrieval can match centralized vector databases on speed or scale. It’s whether a system like that can still produce a shared sense of reality, or just overlapping versions of it that never fully align.

When storage, retrieval, and inference start living across a network instead of inside a system, intelligence stops being a single voice.

It becomes something closer to listening through static and deciding what you believe based on what managed to reach you first.

#OPG $OPG

$SYN $MITO

⚡ THE AI DEBATE IS STARTING IN THE WRONG PLACE.

Everyone is asking which model is smarter.

Almost nobody is asking who owns the intelligence.

That's the part that keeps bothering me.

Over the last year, I've watched developers rush into new AI platforms, build products, attract users, and then suddenly adapt to changing rules they never had a say in.

Not because the tech failed.

Because access changed.

🎭 That's the quiet detail most people overlook.

The biggest battle in AI isn't happening at the model layer.

It's happening at the ownership layer.

A lot of "open" AI today still relies on closed gateways.

You can use it.

You can build on it.

But you don't really control it.

And history shows what usually happens when critical infrastructure ends up in a few hands.

📍 At first, nobody notices.

Then everyone depends on it.

Then the dependency becomes the product.

The risk isn't some sci-fi scenario where AI becomes too powerful.
The risk is waking up one day and realizing the digital intelligence powering everything around us is controlled by a handful of entities.
As agents become more autonomous and AI starts interacting with economies, services, and networks directly, that question becomes impossible to ignore.

Who owns the rails? 🛤️

That's why OpenGradient feels different.

Not because it's promising another breakthrough model.

Not because it's chasing attention.

🌱 It's focused on something more fundamental:
Making intelligence operate like an open network instead of a rented service.

A system where participation matters as much as performance.

Where builders aren't just users.

They're stakeholders.

Most people are watching the AI race.

👀 Very few are watching who gets to write the rules.

And that might end up being the more important story.

Two futures are forming in parallel.

One where intelligence is permissioned.

One where intelligence is native to the network.

Choose carefully.

The model is what people see.

The ownership layer is what they'll eventually live with.

#OPN $OPG @OpenGradient

$SYN $SPCXB

I kept seeing OpenGradient mentioned in conversations around decentralized AI, so I spent some time looking into it.

What surprised me wasn't the AI part.

It was what sits behind it.

Most people talk about decentralized AI like it's a race for more compute, more GPUs, or cheaper inference.

Fair enough.

But after using AI tools every day, I've started noticing something else.

The real value isn't always in the model.

It's in the context.

The little things it learns over time.

How you write.
What you ignore.

The questions you keep coming back to.

That's the stuff that quietly compounds.

And it got me thinking: if AI becomes a permanent layer in our lives, who actually owns that accumulated context?

That's where OpenGradient feels different to me.

The project isn't just asking where AI runs. It's asking whether the intelligence built around you should stay locked inside a platform in the first place.

Most people overlook that question because it's not as exciting as new models or benchmark scores.

But it's probably the one that matters more.

Anyone can rent compute.

Not everyone can recreate months or years of learned context.

Watching this space evolve reminds me of early crypto.

Back then, ownership meant assets.

Then it became data.

Now we're slowly moving toward ownership of intelligence itself.
Not artificial intelligence.

Personal intelligence.

The digital version of everything your tools learn about you over time.
That's the quiet detail I keep coming back to when I look at @OpenGradient .

The infrastructure is interesting.

The question underneath it is even more interesting.

#OPG $OPG

$JTO $ADX

Crypto is obsessed with the entry.

Which token. What price. What size.

Everyone has a framework for buying.

Almost nobody has a framework for what happens after.

I realized this a few months ago.

I spent nearly two weeks researching my last Bitcoin entry.

Charts. On-chain metrics. Macro trends.

The full process.

Then the BTC arrived in my wallet.

And I spent exactly zero time deciding what it should do next.

It just sat there. Like most positions before it.

I never considered that a problem.

The purchase felt like the decision. Everything after was just waiting.

But waiting is also a decision. It simply feels passive enough that we rarely think about it that way.

If 0.25 BTC sits idle for eight months while opportunities offering around 4% yield are available, that's roughly $140 in potential value left unused.

Not a huge number.

But over time, small gaps become meaningful.

That's the idea behind Bedrock 2.0.

uniBTC acts as a unified access point, routing Bitcoin capital across institutional-grade strategies, delta-neutral vaults designed to generate returns regardless of market direction, lending markets, credit markets, and RWA opportunities beyond typical crypto cycles.

BRclaw serves as an AI-powered analyst, helping users understand allocation options without needing a deep quantitative background.

The infrastructure already exists.

More than 5,000 BTC has been deployed, with $382M in TVL across 15+ chains.

The challenge isn't access.

It's having a plan for capital after the buy button is pressed.

The edge in the next cycle may not come from finding a better entry.

It may come from making your holdings work harder while you hold them.

How much time do you spend planning the purchase versus planning what your capital does afterward?
😎

#bedrock $BR @Bedrock
$RIF $TAO

I had a conversation recently with someone who has been holding Bitcoin since before most people around him cared about it.

He did not get lucky.

He was patient.

He sat through bear markets.

Ignored the headlines.

Stayed consistent when everyone else seemed to be changing their minds.

By almost any measure, he succeeded.

But the part that surprised me came later.

He told me that the hardest question he faces today has nothing to do with price.

Not when to sell.

Not when to buy.

Not even where Bitcoin goes next.

His question is much simpler.

What should a long-term Bitcoin holder actually do with their capital between now and then?

For years, there was no real answer.

Bitcoin was treated as a passive asset.

You secured it.

You held it.

You waited.

That was the entire playbook.

Today, the landscape looks very different.

The tools around Bitcoin are becoming more mature.

Not just trading tools.

Actual capital infrastructure.

Strategies designed to generate returns without relying entirely on market direction.

Access to lending and credit markets.

Exposure to opportunities that are not tied to the usual crypto cycle.

Systems that help holders make decisions based on risk and objectives instead of guesswork.

That is why I started paying attention to Bedrock 2.0.

What stands out is not a future vision.

It is that the foundation already exists.

Thousands of BTC are already being deployed across multiple chains and strategies.

The conversation is gradually moving beyond simple custody.

Toward capital efficiency.

Toward portfolio construction.

Toward making Bitcoin productive without abandoning a long-term thesis.

My friend still has not settled on his perfect answer.

But I think the discussion itself says something important.

The question is no longer whether Bitcoin can become a more active asset.

The question is how many holders are still thinking about it the old way.

Have you started looking at Bitcoin differently?

#BEDROCK @Bedrock $BR
$D $ID

#BinancePickAndWin
What stands out to me about #BinancePickAndWin is that it transforms attention into participation. Instead of simply consuming content, you're encouraged to think, decide, and take ownership of your prediction. That shift from observer to participant creates a deeper connection with the experience. Win or lose, there's satisfaction in knowing the outcome reflected your own judgment. In a time when many online activities feel passive, anything that rewards engagement, confidence, and thoughtful decision-making feels refreshing.

The beautiful game doesn't lie.

Form, momentum, squad depth these aren't just stats, they're signals. While the crowd bets on names and jerseys, the sharp minds read the pitch, the press, the fatigue in a fullback's legs by the 70th minute.

That's why I don't just watch football.

I study it.

Every week, #BinancePickAndWin reminds us that prediction is an art and those who respect the game, respect the data, and stay disciplined? They win more than just bragging rights.

Drop your pick below 👇 — who's dominating this weekend?

⚽ Trust the process. Back the analysis. Own the result.

I wasn't thinking about capital markets.

I was just trying to understand what serious money does with Bitcoin when it stops trying to trade it.

One thread pulled me in.

Then a dashboard.

Then a conversation that made me put my phone down for a second and actually think.

Because somewhere between the yield discussions and the collateral frameworks, something clicked.

I thought I was watching people optimize Bitcoin returns.

But I was actually watching a financial system being built from scratch.

Think about what's quietly happening right now.

BTC is becoming collateral.

Not metaphorically.

Literally.

Credit is being extended against it.

Real-world assets are starting to settle around Bitcoin liquidity.

Structured products are forming that don't reference Bitcoin as a bet they use it as a foundation.

That's a different world than we were living in three years ago.

A few years ago Bitcoin was the thing you held while the system figured out what to do with you.

Today it's becoming the thing the system gets built around.

That's not a small psychological shift.

That's a complete inversion.

Projects like Bedrock sit inside this moment in an interesting way.

Not just as a yield tool.

But as early infrastructure in something that doesn't have a name yet.

Because when Bitcoin becomes real collateral, you eventually need risk desks, credit pricing, liquidity depth, and yield curves. You start asking who becomes the Goldman Sachs of BTCFi. And you realize that question is serious now in a way it simply wasn't before.

The hardest part isn't technical.

It's accepting that Bitcoin might be transitioning from the asset we protect to the foundation everything else gets built on top of.

And if that's true, we're not early to a trade anymore.

We're early to an entire financial architecture.

So the question I keep sitting with is this when Bitcoin becomes the base layer of its own capital markets, who actually understands what they're holding?

@Bedrock $BR #Bedrock

$STRAX $STG

Last month, I found myself staring at my Bitcoin wallet longer than usual.

Not because the price moved.

Not because of the news.

But because it was just sitting there.

For years, that was the point.

→ Buy.
→ Store.
→ Wait.

Patience was the strategy.

Discipline was the edge.

Then I fell down a thread about BTCFi.

That led to a dashboard.

Then to a discussion about using Bitcoin in ways I hadn’t seriously considered before.

Eventually, I came across BEDROCK, positioned not as a revolution, but as part of this quiet shift toward making idle Bitcoin do something.

I thought I was being responsible by doing nothing.

But I realized I was actually avoiding a new kind of decision.

A few years ago, holding Bitcoin felt almost ideological.

⇨ You were early.

⇨ You endured volatility.

⇨ You believed in scarcity.

Today, the conversation is different.

It’s less about survival and more about utility.

Holders are starting to ask: if capital can move elsewhere and generate yield, why should Bitcoin remain untouched? Not because they’ve lost conviction.

But because expectations have evolved.

BTCFi isn’t just a technical layer.

It’s a psychological one.

It challenges the identity of the “patient Bitcoiner.” It introduces a subtle tension between purity and productivity.

Between long term belief and short term optimization.

The opportunity no longer demands courage.

It demands participation.

And that changes things.

Because once your Bitcoin becomes productive, it stops being just a store of value.

It becomes a tool.

And tools invite experimentation.

I’m still holding.

But I’m holding differently now.

More aware.

More curious.

Slightly less passive.

So I wonder is this the natural evolution of Bitcoin ownership, or are we slowly rewriting what it means to be a true holder?

@Bedrock #bedrock $BR

$OSMO $SENT