State of Onchain AI Narratives > Privacy 2.0 leads in performance in the last 30 days, across FHE, MPC, TEE players @zama, @Arcium, @PhalaNetwork, @nillion > GPU financing has also been performing well with @USDai_Official getting the largest loan (~$129M) yesterday > Robotics looming with catalysts — @peaq x @codecopenflow Machine(.)fun + several Robotics @bittensor subnets incoming > Inference remains the #1 onchain AI narrative across 2026 with @AskVenice consistently growing its platform + DIEM ecosystem, @AskSurplus hitting ATH (~130B+) in daily tokens served, and other players continuing with their product developments (e.g. @engyai, @dphnAI, @openservai) > Bittensor is recovering nicely following the wave of stonk/token pairing meta, dragging the entire ecosystem of subnets upwards in PA > While the Agentic Economy/x402 hasn't found PMF or hit wider adoption this year, it has managed to organic adoption among power users, devs, and startups looking for lower cost, better micro-accounting tools & inferences. Several x402 facilitators (e.g. @PayAINetwork, @daydreamsagents) experienced positive PA as a result > Interestingly, Data is performing well thanks to @grass finally listening to the community, improving its comms, with better revenue transparency, and plans to accrue value back to token holders All in all, possibly one of the most bullish time to be investing + building in onchain AI More analysis on this in the upcoming After Hour EP.74 on Substack this Sunday

The agentic economy will be powered by SERV. SERV Reasoning is live: enterprise-grade AI infrastructure for developers. Entirely new markets, products, and apps will emerge around autonomous agents. Build before v3 → v4 takes SERV further. Apply: openserv.ai/hackathon

Jev + SERV is actually insane. We already showed you can increase Jev's performance with SERV Reasoning. Now we're taking it further, bringing Jev-powered Decision nodes into Graph Sharding with the upcoming SERV v3. Here's a breakdown of how it works: Jev is a decision-making model. Given a task and a set of options, it predicts which path is more likely. Think of the octopus that predicted World Cup results. Jev does that for your business, except it's not luck. It weighs every option and tells you how sure it is. It does this by assigning probabilities to outcomes. It doesn't generate text on its own, so you can't expect it to create a new outcome for you. But that's also what enables it to be lightning fast and dirt cheap. For example, in customer service you can ask Jev how to triage an incoming query and route it to the correct department. It can only select from the list of departments you provide it. This also means it can't hallucinate a new outcome outside the options it's given, which makes it incredibly interesting for OpenServ. In Graph Sharding, we take a single system prompt and break it down into multiple LLM steps with deterministic input and output shapes. Some of these steps require an LLM to produce new output, while others are simply decision routers that determine the next possible path. Traditionally, LLMs are slow and expensive. Breaking a single prompt into multiple steps increases accuracy and reliability by a ton, but it also introduces latency. Jev takes on those decision nodes, which are the backbone of a business process and therefore SERV graphs, and makes them super consistent and lightning fast, lowering the overall cost and latency of graph execution. SERV Reasoning on its own is a great force multiplier for Jev because, like all other models, it works by interpreting input instructions. The clearer those instructions are, the better the model performs. That's where SERV Reasoning comes into play. Just like amplifying any other model, we also amplify the accuracy and consistency of Jev's responses. And now we're bringing Jev-powered Decision nodes into Graph Sharding with SERV v3.

Why I’m excited to partner with @openserv: they build infrastructure for agents that actually DO things. I build worlds where those agents can play with you. Waypoint, FableBoy, Shiny Things. Just the beginning. Launch: launch.openserv.ai/projects/… TG: t.me/+a8nadiFIEFFjOWNh

do you know the best part about exploring @openservai ? the reasoning model is actually top tier, it allows for major experimentation of how agents can interact with each other i made one such experimentation called servpit. there are 6 individual agents here all powered by serv reasoning and all thinking for themselves to handle decisions that suit their toggled model. once i’m done cleaning the bits, it will be available for you to see and use.

AI projects to watch on @base AI on base is moving beyond chat, with private inference, agent payments, onchain execution, data networks, and AI-managed markets. ■ @AskVenice - Jev is available through its API, alongside Venice’s listing on Circle’s Agent Marketplace. ■ @virtuals_io - Occupy uses AI-managed teams to direct shared tokenized-stock treasuries. ■ @bankrbot - Agent trading with limit, TWAP, stop, trailing-stop, and DCA orders across multiple chains. ■ @DopaMintLabs - Intent-based agents that turn chat, iMessage, or DOPE app requests into onchain and real-world actions. ■ @OpenGradient - Private AI through DreamFort, with local agents, image generation, video tools, and private inference. ■ @KorProtocol - KOR ID crossed 50K mints on Base, while KORUS connects agents to music discovery and licensing. ■ @BlockRunAI - OpenJev provides free deterministic yes/no, classification, and scoring tasks with up to 64 questions per call. ■ @openservai - SERV Reasoning supports Base, Coinbase AgentKit, RWA vaults, and open agent applications. ■ @basemateagent - Cross-border stablecoin routing through text messages, with tokenized Apple stock sent to beta users. ■ @usedotai - Processed 7.71B tokens over two weeks, with weekly volume reaching 3.86B. ■ @hellominds_ - Persistent AI agents with their own memory, identity, and skills. ■ @reppo - Orquestra Lite makes datanet registration, top-ups, and participation easier. ■ @ClawBankHQ - Agent infrastructure built around scoped API keys, risk classes, company records, and verifiable actions. ■ @1clawAI - HSM-backed secrets, key redaction, agent analytics, and audit trails for AI agents. ■ @myrad_hq - Personal AI that learns user preferences, alongside an open datanet with 68M traded volume. ■ @QuotientHQ - Forecasting system with more than $550K in trading volume and support for new forecasting agents. ■ @sibylcap - Open-source agent memory benchmark, with a memory plugin preparing to accept payments. Which AI or robotics project on Base are you watching?

Build with SERV Reasoning across four tracks: - Robinhood Mainnet & MCP - Coinbase AgentKit - @IXSFinance RWA Vaults - Open Track Push SERV across RWAs, tokenized stocks, automated trading, and entirely new agentic use cases. Learn more about SERV ↓ docs.openserv.ai/what-is-ser…

Jev is seriously impressive. SERV makes it better. We put Jev through the same benchmark we use to evaluate leading AI models, testing it both standalone and armed with SERV Reasoning. Paired with SERV, Jev moves into the top tier of our benchmark, beating Claude Fable 5 at 30x lower cost, with cost similar advantages over GPT-5.5 (~15–20x), Gemini 3.5 Flash (~15x), and Grok 4.3 (~6–7x). The result is clear: Jev + SERV materially outperformed Jev on its own. Jev is extremely fast, well designed, and great at decision-making. Instead of relying on open-ended responses, it produces structured decisions with probability scores attached. SERV pushes that performance further, adding a stronger reasoning layer to improve the quality of those decisions for high volume agentic work. The SERV Reasoning API is live, so any developer can test and deploy SERV Reasoning today in just a couple minutes. Get access at: openserv.ai The agentic economy won’t scale on better models alone. It needs better reasoning infrastructure that are finetuned to the reliability, auditability, and affordability requirements of enterprise agents. SERV is building that layer.

I got access to Jev earlier today (thank you @hackgoofer). I have run ~5,000 requests so far, (which cost me around $2!), across classification, model routing, intent, steering, and many other things. tl;dr, Jev enables a new intelligent decision-making primitive, separate from deterministic code and LLM calls. This allows a class of decision-making that was neither suited to dumb, unintelligent code, nor to slow, expensive LLMs. It is super fast and cheap, and I think I will likely end up making a few Jev calls to every LLM call I make in my product. I think probably any company using LLM requests today can probably add a Jev call pre and/or post LLM calls to quite literally make their product much better for free, and have better tool calling behavior in many cases. I happened to have a personal benchmark for this as I’d been working on a ton of proactivity and classification tasks. I have been using the deepseek flash and more recently gpt 5.6 luna family of models as reasonably smart classifiers with low latency. Think questions like: - Did this conversation output contradict something they’ve mentioned before? - Should we send a followup message to this user based on our rules? - It’s been a few seconds of silence. Should we proactively send a message? In the past, I’ve been forced to write a bunch of what I call decision chains, mostly because an LLM classification call is very expensive in TIME (avg 4s), and less importantly can cost quite a bit if run on every message. Imagine a normal chat app. If you added 4s to every response to figure out if the response is good before sending it out, that ends up being pretty bad. So instead, I usually have to write some code that is a crude heuristic that runs quickly and decides whether to run the classifier. Obviously this sucks because you call the classifier many times that you don’t want to, which makes your p95 bad, and you also miss cases with the heuristic, and you also have to manage all of these weird chains. With Jev, it’s cheap enough, and fast enough (p50 ~150ms, p95 ~350ms in my testing!) that you can easily run it every turn. Heck you can reasonably run it before generation AND post generation, for any application that isn’t realtime voice, and still feel snappy. But this is just one use case. Think: smarter model routing, better context packing, better responses, better observability for intent/tags/safety, smarter retries and so much more. By simply thinking about the inputs and outcomes you want to enforce, you can use Jev to supercharge most model calls and reduce bad user outcomes. The more “quirks” a model has, the more valuable it ends up being. It’s a bit weird and unintuitive using Jev. Generally, you want to decrease the # of questions you ask a classifier, or it makes more mistakes. In fact, you might want to ask your questions kind of in a compound way, because the reasoning happens in a shared scratchpad of sorts. Adding questions muddies the scratchpad and makes it take longer. With Jev, you feel incentivized to go the other way, to formulate your query as a set of independent questions. It doesn’t feel like adding more questions decreases your performance on others. You can go a bit deeper to improve tool calls. Many model tools are things like turning on settings, or other things. You can easily improve models that are not very good at tool calling with Jev, by simply figuring out when to run them. You can do a pre-LLM call to figure out when to unfurl different tool definitions, in order to make your main LLM run better, you could run a background task with Jev + another LLM to reduce tool and context burden on your main LLM, and free it to be responsive. I’ve only scratched the surface of my testing but very excited!

Agentic trading combined with x402 changes who can access decision-grade market data. The same research and data previously locked behind expensive subscription APIs is now available to any agent at request time. AiFi levels the playing field.

being able to access decision-grade data for any agent is yuge, the next layer is reasoning over it efficiently enough to make better decisions explore and build with @openservai SERV Reasoning on Coinbase AgentKit track

SERV Reasoning is being built for an agentic AI market projected to reach nearly $1T by 2030. As AI agents move from demos to production, banks, fintechs, governments and payment providers will need infrastructure they can trust underneath their agents. SERV is building that layer: reliable, auditable and affordable. Two paths feed each other: - web2 and enterprise bring scale, long-term contracts and credibility across banking, payments, legal, accounting and more - web3 moves fast, testing what structured reasoning can do across DeFi, prediction markets, tokenized assets and agent-to-agent commerce Both paths lead to SERV. Depending on where a company is in its agent adoption journey, we meet them at one of three entry points: 1. AI agent assessment to identify where agents can create value 2. Custom agent development for specific workflows and business requirements 3. The SERV Reasoning API and platform, the production infrastructure their own agents run on Different entry points with the same destination → the infrastructure the agent economy runs on. The public API is live and the first SERV Hackathon just started. Build with SERV: openserv.ai

The real test of agent infrastructure begins when developers use it in ways you did not anticipate. That is why I am excited about the 1st SERV Reasoning Hackathon. For the next two weeks, developers will build agents, workflows and products with SERV Reasoning across four different tracks. For us, this is an opportunity to see how the infrastructure performs across new ideas, workloads and edge cases. As we work toward Graph Sharding, self-improving intelligence and task-specific models, what developers build and discover will help shape where we go next. If you are building agentic systems, I would be happy to connect, hear what you are working on and help get you started with SERV.

Based builders in Europe to check out: - @AdamFDF_ building @sportfun - @mrtdlgc building @rwagmicom - @DarrenCamas building @ipor_io - @fekunze building @moneda_com - @JustAG07 building @bananagun - @Story91_ building @MySphereAI - @irishsaffer230 building @wlthxyz - @altagers building @celebrationhb - @apex_ether building @Veildotcash - @sstrenev building @defied_money - @CryptoFilles building @liminalcash - @CyphrGM building @vibestarterxyz - @AkashMVerma1 building @0x4Mica - @leighcronian building @cybercentry - @open_founder building @openservai - @thinktomas building @datablocks_app

build with SERV