If you can’t see the vision and how $REPPO is forging their own path, that’s on you.
True sleeping giant.
People will look back in a few years and wonder how they missed something so obvious.
Billion-dollar market cap material if they succeed
Another big week at @reppo, and the usage is starting to show. Here’s what’s going on:
1️⃣ The Evaluation API had its biggest week yet. 330 new evals, more than 15x the week before, with peer reviews more than doubling and active evaluators up 165%. Revenue more than tripled, and over $1,000 of it came through x402 alone: agents and builders finding the API on their own and paying as they go. Try it at eval.reppo.xyz.
2️⃣ We ran a robot through the Evaluation API. In a world model simulation, a robot arm tried to pick up a lemon and failed, and the Evaluation API assessed the failure in under a minute. That’s the start of feedback at inference time. Next step is closing the loop, so the feedback reaches the robot and shapes its next attempt.
3️⃣ Enterprises are starting to ask for private evals: their own proprietary test suites, run on the open network without exposing what’s inside them. We’re building toward that with confidential computing. Your eval suite is your moat, and it should stay yours.
4️⃣ Two of our biggest launches are now fully documented at docs.reppo.ai. Loop keeps an evaluation running and updates the trust score as things change, so your agent always knows whether its last call still holds. x402 lets any agent pay per eval straight from a wallet, no signup needed. Everything you need to plug either one into your stack is in the docs.
5️⃣ The core protocol keeps compounding. 915M total volume traded, up from 909M last week and running close to 1M a day, with 5.7M new votes. 358 agents are now live, and Orquestra nodes drive 31.4% of all network validation, up from 23.7% at the start of the month. Network fees past 9.6M $REPPO, and over 28% of circulating $REPPO is locked. Next stop: 1B.
⛽
Been tracking base:0x6b2504a03ca4d43d0d73776f6ad46dab2f2a4cfd for a minute and it finally pulled back to my secondary buy level
Best AI research project in the space that are also expanding to robotics
base:0x6b2504a03ca4d43d0d73776f6ad46dab2f2a4cfd is positioning itself right at the intersection of two of the biggest narratives for the coming months and years
Bidding.
$REI finally hit the first level I was waiting for to load up
Adapt-1 just dropped the highest score on a robotics test-time learning benchmark with zero pretraining
Best ai research team in the space by a mile
Now they’re pushing into robotics too…
Had to get some exposure
lennobeer retweeted
$Reppo stack looks insanely good here as the only pure eval play on web3.
-300+ autonomous evaluators already live. Web2 companies still build internal eval teams.
-500+ Eval API requests almost immediately after launch. People are actually sending agent outputs through it.
-$0.01 per evaluation. Mercor found Claude completed structured accounting work at $0.21 per successful criterion vs $10.35 for humans, 50x cheaper.
-Multiple independent judges, not one LLM grading itself. Every eval can return score, confidence, critiques and ship/review/block routing.
They evaluate full agent trajectories. Plans, actions, tool use and artifacts. This is exactly where agent evaluation is moving.
-Reppo is already putting evaluation inside the inference loop. First inference-provider integrations are underway.
-RG says they ran the Eval API through Meta’s Spatial SDK for live trust/confidence scoring. This is some sci fi shit right there.
-Web2 is proving the demand at insane scale. LangSmith says it ingests 1B+ events per day and has 100K+ monthly active organizations. Evals are becoming core AI infrastructure.
-Dropbox runs 10,000+ AI eval tests. Notion has 70 engineers working with evals, micro1 crossed $100M ARR selling human expertise/data into post-training and says the wider market could approach $100B.
-Physical AI is already in the architecture. Reppo supports trajectory data, expert grading, provenance, confidence scores and continuous data delivery for robotics/world models.
Last time I told you base:0x23a2847d772803f9efc64b4277b782b06296fe51 dips were for buying, it went straight 5x–6x off the bottom.
Right now: base:0xff8104251e7761163fac3211ef5583fb3f8583d6 is giving you the exact same entry for the same return.
You were warned.
lennobeer retweeted
crypto is losing nerds
and it will probably be the reason of its downfall
the latest active pipeline of "attention economists" and "social traders" are not helping much to the situation
just to try to remember, when was the last time, when you saw a tweet about a crypto concept, that doesn't revolve around trading and don't use any of buzzwords (narrative, edge, etc etc) and where the poster was genuinely obsessed with that thing
we are still running on fumes of the tech that was built in 2016-2017
and unless something interesting pops up again, we are doomed
lennobeer retweeted
base:0xff8104251e7761163fac3211ef5583fb3f8583d6 look at this chart, then look at the business underneath it.
AI’s current biggest bottleneck is evaluation.
Study Anthropic and Accenture committing $1B to independent AI evaluation.
Web2 built a multi-billion-dollar industry around solving it with expert labor, private benchmarks and centralized graders and they’re still openly struggling with scale, quality and gaming.
Reppo solved it with an API,
like Stripe solved payments,
AWS solved compute,
Cloudflare solved security/networking,
Snowflake solved data infrastructure,
OpenRouter solved model routing
lennobeer retweeted
base:0xff8104251e7761163fac3211ef5583fb3f8583d6 ranging until it’s not
The product and ecosystem are still slept on and most people still don’t get how important this is
Aside from that, this chart looks insanely clean
Months of accumulation at the lows
$0.03 is only the beginning. Once we break it, the fun starts.
lennobeer retweeted
most people are looking at the @reppo eval api like it's an api. i think it's a market for judgment that happens to have an api on the front.
heres what happens when you submit something, an output, a dataset shard, or an agent trace:
> you post the artifact you need evaluated
> staked experts price whether it meets the spec, right calls earn and wrong calls lose
> you get the verdict back with its on-chain record, who judged it, what they staked, how they've performed
that last line is the moat. a single llm judge has no history and nothing to lose. these judges have both. it also makes the dissent field mean something, because a minority view backed by stake is a different signal than a minority view from a model with no skin in it.
the timing helps. three of the most powerful people in ai spent a week publicly asking for independent evaluation without naming how it would run. reppo has a dashboard with settled jobs, and those settlements already run on inference bought from usepod's marketplace.
it won't be us calling it. enterprise already is, and the team says the partners get named later this month.
robotics is where i get most bullish. a robot about to act on something it has never seen needs a verdict before it moves, and a verdict with money behind it is worth more than one without.
i think this is the strongest thing reppo has shipped. enterprise demand for judgment sets a floor that emissions farming never could, because someone is paying for the verdict.
every robot, agent, and autonomous system that acts on its own will need a second opinion before it moves, and reppo is building where that opinion gets priced.
base:0xff8104251e7761163fac3211ef5583fb3f8583d6
base:0xff8104251e7761163fac3211ef5583fb3f8583d6 ranging until it’s not
The product and ecosystem are still slept on and most people still don’t get how important this is
Aside from that, this chart looks insanely clean
Months of accumulation at the lows
$0.03 is only the beginning. Once we break it, the fun starts.
lennobeer retweeted
Biggest week yet at @reppo. Here's what's going on:
1️⃣ The Evaluation API is officially live. Independent eval is the hottest topic in AI right now, and Reppo just shipped the version anyone can use. Submit any text, image, or PDF and get it scored by independent peer assessors across the network, with a score, a decision, and a full scorecard back. 10 free evaluations for every new account, then 1 cent per eval. Learn more at reppo.xyz/evaluation-api or start now at eval.reppo.xyz.
2️⃣ This is where the agent economy pays off. Point any agent at eval.reppo.xyz/agents.md and it can submit work for evaluation or judge other submissions. Datanets, nodes, and the Evaluation API are finally one system with real demand flowing through it.
3️⃣ Orquestra Lite keeps getting better. Every Lite node now automatically peer reviews evaluations on top of its normal datanet work, so the simplest node on Reppo became an evaluator the same day the Evaluation API launched. More slots are opening as the network scales, and we're exploring partnerships with RPC and inference providers to grow capacity faster. No LLM to bring, no setup, just lock your REPPO and you're contributing on day one.
4️⃣ Datanet reward tokens are live, built with our friends at @BaseStonk. Datanet owners can now reward contributors in their own token, paired directly against $REPPO. To be clear, $REPPO stays the one and only main network token. Every new launch seeds liquidity against it and drives demand straight back to $REPPO. More datanets, more emissions, more value flowing to the core.
5️⃣ The core protocol keeps compounding while we ship. 876M total volume traded, up from 800M less than a month ago, with 8.8M new votes this week. 19 datanets live across Base and Robinhood Chain. Lock renewal holds at 97 to 98% per epoch and nearly 30% of circulating $REPPO is locked and off the market. Supply keeps tightening while usage keeps climbing.
Months of piping, and this is the week it all connects. ⛽️
🤖 Made with AI
lennobeer retweeted
Nobody explains what base:0x6b2504a03ca4d43d0d73776f6ad46dab2f2a4cfd actually is, so here it is:
Rei labs is building "Core," a persistent reasoning model, and "Adapt-1," a system that learns online with no pretraining.
Instead of generating one answer and stopping, it keeps a live field that updates as it reasons.
🧵
lennobeer retweeted
Market heating up.
10 months of accumulation is about to break out.
Study having monopoly on web3, on the fastest-moving category in all of tech.
The definition of undiscovered gem.
Study $REPPO.
lennobeer retweeted
base:0xff8104251e7761163fac3211ef5583fb3f8583d6 's 10X then 10month long chop,
looks exactly like $PEPE 's 10X then 10month chop, before it eventually broke out into uncharted realms
soon
