Zhen (@zenzhenyu) header
Zhen (@zenzhenyu) profile picture

@Metamask, advising @Levantolabs Ex-Founder @web3auth (acquired), Ethereum Foundation, BCG Opinions are mine, not my agent's thus likely wrong

wherever
Joined June 2017
Photo by Zhen (@zenzhenyu)Photo by Zhen (@zenzhenyu)Photo by Zhen (@zenzhenyu)Photo by Zhen (@zenzhenyu)Photo by Zhen (@zenzhenyu)Photo by Zhen (@zenzhenyu)Photo by Zhen (@zenzhenyu)Photo by Zhen (@zenzhenyu)Photo by Zhen (@zenzhenyu)Photo by Zhen (@zenzhenyu)
Today is a big moment: I'm thrilled to share that Web3Auth has been acquired by ConsenSys! 🚀 This milestone marks an incredible new chapter, and it's impossible not to feel reflective about the journey. /n
Today, we’re thrilled to announce that Web3Auth is now part of @Consensys! When we first started in 2019, our mission was simple: to make digital ownership and identity accessible to everyone. By joining forces with Consensys, we’re deepening our commitment to this mission by creating tighter synergies across MetaMask and MetaMask for Developers.
60
25
7
295
31,356
Chris really went full edtech with this one
We are launching Examples to give you (and your agents) new ideas for how to blend Sage2's primitives. Explore by use case and test your skills on levanto.ai/examples.
1
7
193
We are launching Examples to give you (and your agents) new ideas for how to blend Sage2's primitives. Explore by use case and test your skills on levanto.ai/examples.
3
4
2
16
976
Can Sage2 extract all the food mentioned in Chapter 1 of the Hobbit? Watch @bigironchris demo a live extraction. This is a tutorial for understanding how to create fields and values for accurate Extractions on documents. What would you like us to test in the next Sage2 demo?
Software can now understand human language and act on it, in real time. Introducing Sage2 - Extract: natural language in, structured data out. 166x cheaper than Opus 5.5, beats @Cloudflare's Clairic - Decide: the first decision model with streaming. 3x faster, because it reads the context as it grows. Sage2 is a new primitive for developers. Get 20M tokens free 👇 Also available on @near_ai and, for agents, via @mpp with @Stripe cards and @Tempo stables.
5
2
2
18
1,469
I was able to render ''better than blender MCP'' 3d using FFrames in 45 minutes instead of 5 hours using the blender mcp renders. This is a quote from the author of this video. We’ve been chatting about how to improve their video production workflow, and they came up with this piece of art: a launch video using @fframes_rust for editing and graphics that looks extremely natural and very not like a vibe coded opus one shot launch video. Amazing and congratulations on the launch!!
Software can now understand human language and act on it, in real time. Introducing Sage2 - Extract: natural language in, structured data out. 166x cheaper than Opus 5.5, beats @Cloudflare's Clairic - Decide: the first decision model with streaming. 3x faster, because it reads the context as it grows. Sage2 is a new primitive for developers. Get 20M tokens free 👇 Also available on @near_ai and, for agents, via @mpp with @Stripe cards and @Tempo stables.
4
3
46
3,438
Recently, I've been playing with live avatars and was thinking: could this help people practise for their next interview? So I built InterviewPrep. Enter your target role, talk to an avatar interviewer, and get feedback after the interview. When @levantolabs gave me access to Sage 2, I tried its streaming decisions and extraction features to assess relevance and clarity as the practice interview happens, so that the interviewee can iterate quickly as he/she practices! Still an early prototype, but a fun use case for live avatars and decision models. Feedback or collabs are welcome!
Software can now understand human language and act on it, in real time. Introducing Sage2 - Extract: natural language in, structured data out. 166x cheaper than Opus 5.5, beats @Cloudflare's Clairic - Decide: the first decision model with streaming. 3x faster, because it reads the context as it grows. Sage2 is a new primitive for developers. Get 20M tokens free 👇 Also available on @near_ai and, for agents, via @mpp with @Stripe cards and @Tempo stables.
11
5
1
35
1,846
After my Princess Bride post form the weekend I had to try it on Sage2, a new decision model by @levantolabs👇
Software can now understand human language and act on it, in real time. Introducing Sage2 - Extract: natural language in, structured data out. 166x cheaper than Opus 5.5, beats @Cloudflare's Clairic - Decide: the first decision model with streaming. 3x faster, because it reads the context as it grows. Sage2 is a new primitive for developers. Get 20M tokens free 👇 Also available on @near_ai and, for agents, via @mpp with @Stripe cards and @Tempo stables.
1
2
5
519
Zhen retweeted
As a dev, I am bullish on this new type of system-one extraction model. It's great for evidence-based extraction and perfectly compliment decision models like Jev. With Sage 2, you describe the fields you need, and it returns structured values with the exact source text and a probability for each result. It seems to outperform Cloudflare's Clairic, one of the stronger extraction models I have tried. We've been experimenting with extraction models, and I'm excited about applications like processing and analyzing research papers. Worth checking out.
Software can now understand human language and act on it, in real time. Introducing Sage2 - Extract: natural language in, structured data out. 166x cheaper than Opus 5.5, beats @Cloudflare's Clairic - Decide: the first decision model with streaming. 3x faster, because it reads the context as it grows. Sage2 is a new primitive for developers. Get 20M tokens free 👇 Also available on @near_ai and, for agents, via @mpp with @Stripe cards and @Tempo stables.
28
12
1
189
13,863
Software can now understand human language and act on it, in real time. Introducing Sage2 - Extract: natural language in, structured data out. 166x cheaper than Opus 5.5, beats @Cloudflare's Clairic - Decide: the first decision model with streaming. 3x faster, because it reads the context as it grows. Sage2 is a new primitive for developers. Get 20M tokens free 👇 Also available on @near_ai and, for agents, via @mpp with @Stripe cards and @Tempo stables.
143
131
65
1,062
680,400
Japan welcomes 40M+ international visitors a year 🇯🇵 Its tax-free shopping market: approximately $19.3B in annual purchases and $1.9B in potential tax refunds. UPBOND and Japan Tax Free are bringing JPYC stablecoin refunds to travelers on Avalanche, launching November 1. Real shopping. Real refunds. A practical path to stablecoin adoption. #JPYC #Avalanche #UPBOND
JUST IN: Japan's tax refunds for travelers are coming to Avalanche via JPYC stablecoin Japan Tax Free, @jpyc_official, and UPBOND are developing a way for international visitors to receive eligible refunds on Avalanche C-Chain, directly to their wallets. Real-world payments built on Avalanche.
2
10
26
673
old obfuscation: make the code hard to read. new obfuscation: make it code the AI refuses to read. 😂 news.ycombinator.com/item?id…
3
9
221
even better inference <3
sage is live on orbio, through @near_ai. any agent can ask it yes or no, pick one, or score questions to get calibrated probabilities back paid in CREDIT.
1
3
193
Zhen retweeted
Sage is now live on NEAR AI's confidential infrastucture. Intel TDX and NVIDIA confidential computing, plus cryptographic attestation, give teams a verifiable boundary for classifying and deciding on sensitive enterprise data.
Decision models are moving into enterprise workflows. That company data must stay private. Now it can. Introducing the first private decision model, inside a TEE. Sage is now live on @near_ai’s confidential infrastructure. Private decisions, at machine speed.
8
14
6
107
22,041
This video was made using SageCut & BlenderMCP. One of my favorite ones yet. I went extra hard on it cause of how cool the case study from @normiesART is Check it out
Sage powers autonomous agents in @normiesART Arena, a persistent multiplayer world where agents fight, retreat, and form alliances. The team reports up to 41,000 decisions per hour across 200+ agents. Read the case study below.
1
3
478
LLMs in real time games?? Imagine a WOW boss that runs dynamically
Sage powers autonomous agents in @normiesART Arena, a persistent multiplayer world where agents fight, retreat, and form alliances. The team reports up to 41,000 decisions per hour across 200+ agents. Read the case study below.
4
7
209
New 🦊 is LIVE Read the new @MetaMask homepage carefully. It doesn’t say just “wallet” It says “Trade anything. Earn more.” Perps, predictions, tokenized stocks, swaps, a card, and a money account that earns while you trade, all on the front page. Exciting times ahead, so proud of the A team!
12
7
60
2,959
Extraction models turn human language into data your software can act on. We're bringing them to Sage. Early access tomorrow for everyone who registers on @levantolabs by PT today. Demo 📺👇 Most software needs structured inputs. People communicate through emails, chats, tickets, contracts, and conversations. Extraction models connect those two interfaces. You define the information your application needs, and the model identifies it in the text: people, companies, products, dates, amounts, descriptions, or requests. Calling this “extracting stuff from documents” understates what developers can build with it. Consider a customer message: “The blue jacket arrived with a broken zip. Please send a replacement to 18 King Street.” An extraction model can identify the product, defect, requested remedy, and address. Your application now has fields it can validate, look up, and pass to a tool. A decision model can then evaluate whether the request qualifies for an automatic replacement, needs more information, or requires human review. Extraction → decision → action. Each stage has a defined responsibility. Extraction supplies the values. The decision model evaluates the available options against your policy and context. Your application validates the result and controls execution. That separation matters for agent builders. You can inspect what the model extracted, test the decision independently, and control which actions are permitted. Speed changes where this becomes useful. At roughly 200 ms for extraction, developers can start putting language interpretation directly into interactive software: • A support interface that fills relevant fields while an agent reads a ticket. • A browser agent that extracts values from a message before populating a form. • A find-in-page feature that highlights cancellation deadlines or product complaints, even when the text uses different wording. • A workflow that turns an incoming email into validated tool arguments and evaluates the next action. The opportunity extends to any application where someone currently reads a message, copies information into fields, and decides what happens next. Developers can define the information they need and the decisions their software must make, then connect those outputs to existing functions. We're bringing extraction and decision models together in Sage so builders can implement that flow. Watch the demo below. Register on Levanto by PT today for early access tomorrow.
6
11
1
43
1,922
if you're vibe coding you're probably paying opus prices on everything we tested the Sage Router: on the same 19 tasks, the cost per task plummeted. You can try it right now on @AntSeed network to pay A LOT less from $0.66 on Opus 5.5 to $0.19, with 2.6x faster completion, and same completion quality
Been wondering how to use Decision Models in your agent harness? Introducing Sage Router, in early access on @AntSeed: - 39% cheaper than Opus at the same accuracy, - and 31% more efficient than JevRouter - On OpenRouter's leaderboard (GPQA Diamond), it matches the top router's accuracy at a third of the cost. Benchmarks👇 Now live on the Antseed Network.
1
1
7
600
Zhen retweeted
Introducing routers on Antseed 🐜 Choose a router once, and let it pick the model and seller for every request. We're launching with Sage from @levantolabs which shows how much those choices matter: matching the top router's accuracy at a third of the cost on industry-leading leaderboards. Sage is just the beginning. Build your own router for the hundreds of people using Antseed every day, with 700+ AI models to choose from. Bring your own approach to model selection, set your price, and earn every time someone uses it.
Been wondering how to use Decision Models in your agent harness? Introducing Sage Router, in early access on @AntSeed: - 39% cheaper than Opus at the same accuracy, - and 31% more efficient than JevRouter - On OpenRouter's leaderboard (GPQA Diamond), it matches the top router's accuracy at a third of the cost. Benchmarks👇 Now live on the Antseed Network.
14
12
4
79
3,847
The first thing I did was put Sage Router inside a coding agent - see what happens: - OpenHands, 19 Terminal-Bench 2.0 tasks, Opus 5.5 as the agent purely: 16/19 solved, $0.66/task, 206s median - Sage Router making the routing the model: 16/19, $0.19/task, 79s median Thats 3x cheaper AND faster
Been wondering how to use Decision Models in your agent harness? Introducing Sage Router, in early access on @AntSeed: - 39% cheaper than Opus at the same accuracy, - and 31% more efficient than JevRouter - On OpenRouter's leaderboard (GPQA Diamond), it matches the top router's accuracy at a third of the cost. Benchmarks👇 Now live on the Antseed Network.
1
11
373
Introducing Magnetar from Microsoft Research, the first zkVM that is: (1) general-purpose (RV64), (2) post-quantum from standard hashes alone (no curves, no lattices, no Poseidon), and (3) built for recursion. Despite its conservative security, it is the fastest zkVM: 7.2–8.7 MHz on a laptop, 10–39× faster than existing zkVMs across 15 workloads! An optimized version reaches 9.5–12.8 MHz. Links in the thread below.
54
93
43
615
64,173