Pinned Tweet
Probably not obvious to most yet but I think compute is the next commodity Wall Street learns to trade.
If you just look you can see gpus already used as loan collateral and rental prices already have indexes.
Just to explain better oil also became a market when a barrel could be measured.
That is why I think compute becomes one when work can be proven.
In my opinion two layers are needed:
Compute you can trust bit for bit, same results on any hardware .
Inference you can prove came from the model you paid for ofc with a market to buy it.
...two chains building exactly that
I am heavily in both...
"buy when there is blood in the streets"
Inference pipe works already proven
2 models atm ( will be more)
8B ~ 634k $ worth
27B ~ 376k $ worth
Of inference with cost at zero or close to zero just bcs miners are already paid via block reward but still people dont get it why i am saying that $TSC is Bitcoin of inference
Caco retweeted
Agentic tokenized finance needs a verification + settlement layer
$TSC is building it at the protocol level. Nobody else has all the pieces on one chain.
wrote up why 👇
So you are telling me this is the Bitcoin of inference and nobody gets it yet? Yes. Today it ran for the first time.
A coding agent got one task than build an animation. It worked 28 minutes alone for 42 rounds with 1.2 million tokens in.
The model did not run at OpenAI it ran on a single mining GPU on this chain. That was same card with mining blocks and answering the agent in one process. Every answer carried a proof bound to a mainnet block so anyone can check which model did the work.
Whole session sold for 16 cents and miner got paid in the chains own coin you can also check tx it is public.
Now the part nobody understands. For that miner that inference and securing the network are the same job. The card is already running and already paid by the block reward which means every customer token on top is extra income with the hardware cost already covered. There is no web2 cloud that can compete with that structure. They all have to earn the GPU back from customers but here the chain pays for the GPU and customers pay for the margin.
16 cents is not a business and we know that but it is proof the pipe works and that is HUGE . $TSC
One day after Qwen3.8-27B went live on mainnet, it is already serving paid traffic: a full coding session, served by a TensorCash miner and sold through @AskTillyAI, an independent operator on the network.
Every request came with a proof of which model did the work.
Caco retweeted
Thousands of GPUs running in a trust-less decentralised network, producing verifiable inference. Not a vision, it is happening today @Tensorcash.
Caco retweeted
For those paying attention, @100xInsight has been uncovering some pretty interesting details surrounding tensorcash:native lately
And when you put them together, an interesting picture starts to emerge.
It starts with @enrico_otta, an early TensorCash contributor who I also highlighted in my recent article.
@100xInsight discovered that back in 2013, Enrico and @hardmaru (David Ha) were actually listed in the same Goldman Sachs announcement of new Managing Directors.
Max Boonen, co-founder of @B2C2Group, was also working at Goldman Sachs during that period.
Now fast-forward to today:
- David went on to co-found @SakanaAILabs, becoming a prominent figure in AI research
- Max built B2C2 into one of the leading institutional crypto market makers
- And Enrico is now involved with AskTilly and TensorCash, exploring how verifiable AI inference could become a tradable commodity with financial infrastructure built around it
Three people whose careers overlapped at Goldman Sachs, now operating in different but potentially complementary areas of AI, traditional finance and institutional crypto.
Whether they know each other personally or maintain professional connections isn't confirmed, but you can probably imagine how valuable those networks could become.
And that's not all @100xInsight discovered
The official @Tensorcash account recently started following @CumberlandSays, one of the biggest institutional crypto market makers and part of DRW.
But here's where it gets even more interesting
Nate George (@nathereum), who has co-led Cumberland's venture capital activities, has reportedly started following @Tensorcash as well.
Then there's @bronwallet, where @100xInsight spotted tensorcash:native appearing with an actual token balance and price display. Interestingly, Bron Wallet also follows @Tensorcash back on X.
Of course, none of this confirms partnerships, investments or institutional involvement.
But considering that TensorCash isn't just building verifiable AI inference, but an entire financial infrastructure layer around it, I find these potential connections particularly fascinating.
We're talking about AI, institutional liquidity, custody and financial settlement. Different pieces of an ecosystem that could eventually become very important if verified AI compute develops into a globally traded market.
Big credit to @100xInsight for putting in the research 🤝
I genuinely think there's a much bigger picture here that most people are still overlooking.
Connect the dots...
Caco retweeted
so let me get this straight
a $20m AI inference blockchain has former Goldman Sachs MD Enrico Ottavian building around it
he's connected on LinkedIn to David Ha, founder of NVIDIA-backed Sakana AI ($2.65bn valuation), and Max Boonen, founder of B2C2 ($2tn+ traded)
now Nate/nathereum, who lists himself as VC at Cumberland DRW, is following tensorcash:native
DRW is a Wall Street trading giant that traded $824bn in ETFs alone back in 2020, with its crypto arm providing liquidity to Goldman Sachs and Nomura
all this around a $20m project. probably nothing
Today i already showed you how this chain issues real assets. Now i will try to exlpain part how they want to trade.
Every perp you trade on GMX or Hyperliquid depends on an oracle. You know that Mango got drained through exactly that door. The oracle is the weak point of all of DeFi.
On this chain a contract can settle on something the chain measures itself every block that is important bcs tells you how much work the network did even how hard it was. Nobody outside tells the chain what happened. An option on network difficulty already expired and paid out from the block header with no price feed at all.
Repo works the same. You lock your coins with a counterparty they give you stables and the script decides what happens if you default. There is no pool, no liquidation bot, no governance vote just one contract with two parties and the chain enforces it.
Important is that nobody sees your deal. On Ethereum the MEV bots see your trade before the block. Here only a hash of the contract goes on chain. The terms appear when it settles and you can settle privately before that also.
That is the bilateral side and live today.
The next step is a native DEX and the design paper is already public. Built so frontrunning has nowhere to go also with fees and staking $TSC in the chains own coin. This is not live yet so i treat it as a roadmap not a feature yet.
For inference this is very important. A datacenter hedges its output. A buyer locks its cost, it is private until settlement happens still enforced by the chain with no desk in the middle. That is example how oil trades with $TSC is way how compute can trade on $TSC
👀
Said it before and will happy repeat again this chain is a tradfi back office for inference. Then this morning it shows up in a wallet built for family offices and institutions.👀 Let me show you what those people see when they look under the hood.
Every token you ever bought on Ethereum is a smart contract somebody wrote. Most of the hacks you remember like bridge drains, rug through a mint function or hidden admin key lived in that code.
On this chain there is no contract and asset has a fixed format that every node reads same way it reads a normal transaction so there is no contract to drain. First one is already live, wrapped USDC from Base, sitting on the base layer.
Now the part tradfi actually cares most about. The issuer can write into the asset itself who is allowed to hold it. For example only verified wallets and the chain checks the proof itself. BUIDL and $ONDO do this with a contract and an off chain list but here it is native.
And an asset can have child assets also. One issuer for a series of bonds or a treasury that pays yield to holders. Also important rule changes by holder vote and still all native without anything deployed.
Why does this matter for inference? Because when datacenters sell compute ahead and funds finance GPUs they all need instruments not memecoins hahah ( maybe later some for degens hahah). They need instruments, shares, bonds, fund units, with legal terms attached.
The tools are already live and therw is much more will touch it in other tweets. What the chain is waiting for is the first real issuer and the first real buyer of inference which will come through @AskTillyAI > InferenceFI so keep eye on it bcs we at multicycle bet and things will become only better.
刚才Bron Wallet 里居然搜到 tensorcash:native 而且还显示价格...要知道 TSC 还只在Safetrade上列出..太奇妙了...
@bronwallet 是一款专门面向机构、家办和高净值人群的机构级移动钱包。他们今年宣布集成的新项目有 Arc Hbar Stellar Pearl Gonka 这些无一不是具备机构级投资背景的项目。
TSC 作为独立L1，能被 Bron钱包主动且悄悄的集成...结合早上发现官推关注 DRW 旗下的 @CumberlandSays ，无法不让人浮想联翩啊....期待期待。
顺便说下，现在终于有个安安心心存 TSC的地方了... 金豆子必须存在安全的地方，独立管理的IPHONE+Bron Wallet让人放心。
Caco retweeted
刚才Bron Wallet 里居然搜到 tensorcash:native 而且还显示价格...要知道 TSC 还只在Safetrade上列出..太奇妙了...
@bronwallet 是一款专门面向机构、家办和高净值人群的机构级移动钱包。他们今年宣布集成的新项目有 Arc Hbar Stellar Pearl Gonka 这些无一不是具备机构级投资背景的项目。
TSC 作为独立L1，能被 Bron钱包主动且悄悄的集成...结合早上发现官推关注 DRW 旗下的 @CumberlandSays ，无法不让人浮想联翩啊....期待期待。
顺便说下，现在终于有个安安心心存 TSC的地方了... 金豆子必须存在安全的地方，独立管理的IPHONE+Bron Wallet让人放心。
Every Bron account comes with Power Bank. That’s a $5 gas balance, built in, to be topped up as needed.
No token to hold purely for the sake of moving money. No failed transfers because the tank has run dry.
Just one less thing to think about.
Caco retweeted
突然发现 @Tensorcash 新增关注了 @CumberlandSays
查了下这是 @DRWTrading 旗下的加密交易机构，在机构级做市和大宗场外交易领域，属于绝对的第一梯队。
他的分量可以这么描述：高盛和野村进入加密交易时，曾由 Cumberland 提供流动性，贝莱德 IBIT 的年报将 Cumberland 列为比特币交易对手之一。
Hmmm,事情好像变得越来越有趣了。🤔 tensorcash:native
这周的调研结果，主要摸到两个Eo可能很熟的牛人
@SakanaAILabs 的创始人David Ha和 @B2C2Group Max Boonen
前者是2025年《Time》AI百大名人之一，后者是顶级机构级的加密做市商创始人。
简单说 @enrico_otta 虽然在媒体上很低调，但他在传统金融，新型金融科技，加密货币乃至人工智能都应该有不错的人脉。如果 @Tensorcash 能发展起来，EO完全有能力给 $TSC 推开更大的门。
Everyone on my timeline talks AI this compute that and almost nobody asks where the money actually goes.
It goes to inference but running the models not training them. Out of all the computing power used for AI this year two thirds goes to just answering questions imagine that. So not for building smarter models it is just for running them. That is inference my bros and that is where the money is atm.
Ok, fine now we say a project actually produces inference. We are talking about real models, real answers and proof on every one.( only 1 projects does this on this way tbh ngl).
Then the real question starts. Who buys it? How do you sell it? What is the unit , how is delivery proven, how does a datacenter hedge and who settles the trade? Hmmmm InferenceFI?!?!? Yes
Oil figured this out over a hundred years. A standard barrel, an assay, forwards and clearing. So without that oil is just a black liquid in tank.Crypto devs know how to build the machine but most of them have never seen how a real market is built and it is true.
That is why one detail in this race of AI INFERENCE MANIA matters to me. The guy building the commercial side on this chain spent over a decade at Goldman structuring rates products.Let me explain better so building markets that institutions can actually use was literally his job.
Most compute coins cannot even produce proven inference. The one that can is also the one with a market built around it. That is not a coincidence that is the whole point my friends . $TSC
I keep going down the rabbit hole of this chain and I cannot stop and i will not stop.
Everyone talks about it as an AI mining coin but that is maybe half of the story. The other half is a full financial system already built into the base layer. Then when you look at who is around it becomes very wild really really wild times ahead.
Btw the whitepaper is signed by a Japanese pseudonym you can play smart using chatgpt translate the name and it means potato with date April 1st. Most people stop reading there and think it is a joke bcs they think they have magic power using chatgpt/claude shit (fun fact this chain produces the same kind of AI tokens just from open models and with proof on every answer lol)
But everyone with something in head knows this is just the filter for lazy people.
If you just keep reading it stops being funny very fast you can see ten papers with theorems and proofs in the appendix , Bitcoin Core fork with post quantum signatures on steroids , time proof so nobody can fake how long work took and 442 methods in the node API. The only reference in the main paper is Satoshi 2008.
Nobody writes this much math for a joke makes no sense so whoever built this wanted the lazy ones to laugh and leave i swear man.
Also look at who is building on top of it. The company building the commercial side is led by a guy who spent over a decade at Goldman Sachs running rates structuring. Independent devs run pools and a second verification service. This is not a crypto kid project. Go back to my post where i said same playbook but this time tradfi frontruns lazy retail.
No VC unlock schedule. No airdrop farmers. Almost no exchange listings only our safe place to trade POUW ST. Just bunch people from top of TRADFI building quietly... more coming soon $TSC
I already mentioned there is a project doing for compute what Satoshi did for money so it replacing trust with proof.
Proof is the first step and the biggest one. The whole economy on this chain is built around it but still solved very smart in the way like: the same work that secures the chain is real inference that can be verified and sold. No puzzles, no wasted energy.
If inference becomes a commodity it needs what oil has:
A standard unit - here a million tokens of a named model.
Proof of delivery- here is a replay that proves which model answered.
Forwards - so AI companies can lock compute cost and datacenters can sell capacity ahead.
Settlement and repo without middlemen.
This chain already has most of it. Registered models, replayable proofs, KYC assets, repo and options settled with no oracle even quantum safe from block 0 and DEX against frontruning on the way .
That is not a crypto feature list btw this is a Wall Street back office for a new commodity. That is why I said Wall Street will front run retail this time. Degens are just early if they smart enough...
Also lets be honest. Put simply most compute projects today rent a warehouse. This one is the oil field, the assay lab and the exchange in one place.
$TSC
If you don't believe me or don't get it, I don't have time to try to convince you, sorry.
In 2008 Satoshi gave the world a new form of money built on cryptographic proof instead of trust.
The problem was never the banks alone. It was having to trust a third party you could not check.
Today you also buy AI inference on trust. You trust the provider ran the model you paid for at the precision you paid for or you just hope they didnt swap it for something cheaper. You cannot check unless...
Now lets imagine we could check every inference carrying proof that the right model actually ran. We are not talking about trust we are talking about true proof.
This is not some crazy idea of mine. It exists and it is already running on chain. This is a multi cycle bet and it probably gives crypto for the first time a useful role in a global narrative that is not about money.
So back to topic. Same problem, new market. Proof instead of trust, but for COMPUTE.
Last time retail had more than a decade to front run Wall Street. This looks like the same playbook. Only this time it feels like Wall Street will front run retail.
If you don't believe me or don't get it, I don't have time to try to convince you, sorry.
Caco retweeted
Inference is becoming a massive and rapidly growing market.
As inference costs become a material P&L line item, enterprises will increasingly focus on optimizing cost, performance and value extracted from AI models.
But a fundamental question remains: how do you verify you’re actually getting the model you’re paying for?
TensorCash solves the verification problem. inferenceFI builds the marketplace for open, verifiable inference.
The next frontier isn’t just cheaper inference. It’s trusted, efficient and transparent inference markets.
xitter.cf/Delphi_Digital/status/… Nocktime base:0x9b5e262cf9bb04869ab40b19af91d2dc85761722
Can a blockchain reward miners for more than one kind of verifiable computation?
Bitcoin gives every miner the same basic task: search for a valid hash. This provides a consistent measure of mining power because more hashpower means a greater chance of producing the next block. Hashing works well for consensus even though its output has little value outside the blockchain.
Nockchain replaces hash grinding as the costly mining work with generating proofs over NockVM computation. More proofpower gives a miner a greater chance of producing the next valid block, analogous to hashpower in Bitcoin.
Nockchain extends this model beyond a single kind of work. Its base ZKPoW network rewards NockVM proof generation. A separate AI PoW network lets inference providers reuse existing model computation while competing for PoW rewards. Each has its own difficulty and incentive structure.
For proof-based mining to work, proofpower has to remain tied to fresh computation. A miner's chance of winning should scale with effective proofpower without allowing previous work to be reused or amortized across new mining puzzles.
The protocol addresses this by cryptographically binding each puzzle to fresh inputs. Extending the model into a broader compute market introduces another requirement: there also has to be external demand for the computation being incentivized.
One Nvidia H200 sells for 30k. Logan said it himself.
Right now there is equivalent of ~1,500 H200s mining on Nockchain. Same math he used, just today's work rate: 1.54 ExaMAC/s. That is 44 million dollars of iron.
Now open the other tab. NOCK market cap: 2.28 billion coins times 0.014 = 32 million.
Read it again slowly. The hardware mining the coin is worth MORE than every coin in existence combined.
I seen many things in six years but I never seen miners plug 44M of GPUs into 32M coin. Nobody burns electricity for charity. Iron is the most honest money in this market, it cannot tweet, it cannot shill, it can only vote with the power bill.
And it is voting.
Work rate 2.7x in two days. Chart is on nockscan, go check, dont trust me.
Either 1,500 H200s are wrong, or the price is. $NOCK NOCK