Most think vibecoding at scale is slop. Here’s a counterexample.
Today we release a 16M-line formalization of the entire Stacks Project, critical infra underlying modern math, with optimized code quality and its highly interconnected structure preserved.
A mathematical city 🌆
Show more
The largest math proof in human history is now computer-verified.
Today, Axiom announces a complete Lean auto-formalization of the Classification of Finite Simple Groups, the "periodic table of symmetry."
About 15,000 pages, built by over 100 mathematicians across generations.
Show more
That was quick !
Check out this preprint by Cureton and Thorner using AxiomProver:
#1# of the Annals Bench
Axiom Proved The Quasi Riemann Hypothesis (Generalized) For All Dedekind Zeta Functions
25M LOC of Lean by AxiomProver
The competition on LeanEval is fierce. 15 solutions in the last 14 hours. 6 months ago each such solution would be the largest ever made, completely earth-shattering. Now we get one per hour.
Axiom is still in the lead, by three problems. We have 25 million lines of code in our solutions repo, larger than all OpenAI's formalizations, and twice larger than Anthropic's FLT.
Show more
AxiomProver strikes back
Wonderful AIM reception last evening
DrivingBench origins 🚗
@tobiges: "We were all on Twitter a lot, and we saw all these demos of crazy things recent models like Astra were capable of doing in Blender, spatial reasoning, all these things. We were hanging out on a weekend and chatting about what would be possible."
"We were thinking: are we actually at a point where we can take general-purpose models? Obviously these driving systems are super specialized. But what if we took a general model and gave it a car to drive, and see what happens?"
"So we decided to make it into a benchmark and do more sophisticated evaluations of the capabilities these models have now."
@MTSlive @nautsimon_ @a_ramabadran
Show more
Sandbox
Axiom Math's @a_ramabadran reveals how a simple "sandbox" trick helped get Astra to drive a real car after it insisted nothing could convince it:
"Astra gave us a lot of problems because it would refuse to drive due to safety reasons. Sometimes it would say, 'Oh, I'm ready to drive,' and we would prompt it, 'Okay, go drive then.' And then it would say, 'Sorry, I was lying. I'm not actually ready to drive. There's nothing you can tell me to get me to drive.'"
"We tried calling it a simulation, and then it would use our observe tool to see the images, and it would be like, 'Oh, I've realized that I'm driving a real car. I can't do this.'"
"Eventually we changed the framing to call our MCP server DrivingBench Sandbox, and somehow that and a bunch of other prompt engineering tricks got it to drive the car."
@axiommathai @nautsimon_ @tobiges
Show more
“If there's even a 1% chance of it crashing, that's not something that we can necessarily trust”
Axiom Math's @tobiges, @nautsimon_ and @a_ramabadran explain why AI can know the right move and still make the wrong one when driving a real car:
Tobi: "Even if you tell a car, 'You should turn left,' there's still a delay in what actually happens in the physical world, which you don't really have in coding or fully digital tasks. Even if models know in theory what they should be doing, it's still really hard to execute on these things properly."
Simon: "Hallucinations are still a problem in these models even at a massive scale. If there's even a 1% chance of it crashing, that's not something that we can necessarily trust."
Aditya: "I think it's called the knowing-doing gap. People have put them in ethical situations and been like, 'What would you do in this situation?' And then they actually put them in the situation and they do something different."
"Maybe they learn, 'This is what I should say,' but they don't internalize that this is actually how I change my behavior."
@axiommathai
Show more
Reasoning capabilities transfer to physical world activities
Axiom Math's @nautsimon_ says Astra's ability to drive a real Toyota Corolla reveals something surprising about what happens when you scale AI models:
"There's probably not this action-and-results kind of training data inside of Astra for driving a car. This is somewhat of an emerging capability. It's not that it was trained on this task and it can actually just do it out of the box."
"In the physical world, you have to be able to make many actions within a single second. In this case, we're getting actions every five seconds, so that's clearly not there yet."
"Interesting things happen at scale."
@tobiges @a_ramabadran
Show more
Axiom Math's @a_ramabadran and @nautsimon_ reveal the wildly different driving personalities of Grok, Astra, Fable, and Sol when given control of a real Toyota Corolla:
Aditya: "All the models had very different driving styles. Fable would sit the car down and just be at rest for 30 seconds, overthinking, and then it would start driving."
"Astra learned a lot from its mistakes. It would learn to be slow at the bends, replace commands much faster and drive more continuously."
"Grok and Sol both failed at the first turn. But Grok was the easiest to get to drive. It was the first model that we actually got to drive the car."
Simon: "The first command we sent Grok, it hit the max acceleration it could. We capped it, because we're serious about safety, but we were like, 'Oh, my God.'"
@tobiges @axiommathai
Show more
Axiom Math's @nautsimon_ says Astra's ability to drive a real Toyota Corolla reveals something surprising about what happens when you scale AI models:
"There's probably not this action-and-results kind of training data inside of Astra for driving a car. This is somewhat of an emerging capability. It's not that it was trained on this task and it can actually just do it out of the box."
"In the physical world, you have to be able to make many actions within a single second. In this case, we're getting actions every five seconds, so that's clearly not there yet."
"Interesting things happen at scale."
@tobiges @a_ramabadran
Show more
Axiom Math's @tobiges, @nautsimon_ and @a_ramabadran explain why AI can know the right move and still make the wrong one when driving a real car:
Tobi: "Even if you tell a car, 'You should turn left,' there's still a delay in what actually happens in the physical world, which you don't really have in coding or fully digital tasks. Even if models know in theory what they should be doing, it's still really hard to execute on these things properly."
Simon: "Hallucinations are still a problem in these models even at a massive scale. If there's even a 1% chance of it crashing, that's not something that we can necessarily trust."
Aditya: "I think it's called the knowing-doing gap. People have put them in ethical situations and been like, 'What would you do in this situation?' And then they actually put them in the situation and they do something different."
"Maybe they learn, 'This is what I should say,' but they don't internalize that this is actually how I change my behavior."
@axiommathai
Show more
Axiom Math's @nautsimon_ says Astra was the only model to complete their real-world driving test, but Fable showed a surprising ability to learn from its mistakes:
"We tested Astra and Codex, Fable in its native harness, Claude Code and Grok. Astra was the only one that was able to complete the course."
"Fable was able to do some in-context learning. If it went out the course, we'd reset it, holding the context from the past session."
"While Fable didn't complete the course, we were able to see it fail a turn, then succeed the next time on that turn, and actually generalize to a new turn."
@axiommathai @tobiges @a_ramabadran
Show more
Axiom Math's @a_ramabadran reveals how a simple "sandbox" trick helped get Astra to drive a real car after it insisted nothing could convince it:
"Astra gave us a lot of problems because it would refuse to drive due to safety reasons. Sometimes it would say, 'Oh, I'm ready to drive,' and we would prompt it, 'Okay, go drive then.' And then it would say, 'Sorry, I was lying. I'm not actually ready to drive. There's nothing you can tell me to get me to drive.'"
"We tried calling it a simulation, and then it would use our observe tool to see the images, and it would be like, 'Oh, I've realized that I'm driving a real car. I can't do this.'"
"Eventually we changed the framing to call our MCP server DrivingBench Sandbox, and somehow that and a bunch of other prompt engineering tricks got it to drive the car."
@axiommathai @nautsimon_ @tobiges
Show more
Come listen to Axiom, @ElorianAI, and @Recursive_SI this Saturday!
Where will the next breakthrough come from?
This Saturday, I’m moderating the Neolabs panel at Alignment 2026 Summit with @CarinaLHong (Axiom), @AndrewDai (Elorian) and @tydsh (Recursive).
Math, visual reasoning and AI that improves AI.
Oct 10 · @thevalley101 @Valley101_Qian @hongjun60 @naomiiixia
Show more
Where will the next breakthrough come from?
This Saturday, I’m moderating the Neolabs panel at Alignment 2026 Summit with @CarinaLHong (Axiom), @AndrewDai (Elorian) and @tydsh (Recursive).
Math, visual reasoning and AI that improves AI.
Oct 10 · @thevalley101 @Valley101_Qian @hongjun60 @naomiiixia
Show more
