Leanstral: Open-source agent for trustworthy coding and formal proof engineering

https://mistral.ai/news/leanstral

by Poudlardo • 4 months ago

783 185 4 months ago

mistral.ai

Lean 4 paper (2021): https://dl.acm.org/doi/10.1007/978-3-030-79876-5_37

blurbleblurble• 4 months ago

Truly exciting

andai6• 4 months ago

Trustworthy vibe coding. Much better than the other kind!

Not sure I really understand the comparisons though. They emphasize the cost savings relative to Haiku, but Haiku kinda sucks at this task, and Leanstral is worse? If you're optimizing for correctness, why would "yeah it sucks but it's 10 times cheaper" be relevant? Or am I misunderstanding something?

On the promising side, Opus doesn't look great at this benchmark either — maybe we can get better than Opus results by scaling this up. I guess that's the takeaway here.

DrewADesign2• 4 months ago

It’s really not hard — just explicitly ask for trustworthy outputs only in your prompt, and Bob’s your uncle.

miacycle2• 4 months ago

Assuming that what you're dealing with is assertable. I guess what I mean to say is that in some situations is difficult to articulate what is correct and what isn't depending in some situations is difficult to articulate what is correct and what isn't depending upon the situation in which the software executes.

flowerbreeze• 4 months ago

They haven't made the chart very clear, but it seems it has configurable passes and at 2 passes it's better than Haiku and Sonnet and at 16 passes starts closing in on Opus although it's not quite there, while consistently being less expensive than Sonnet.

andai6• 4 months ago

Oh my bad. I'm not sure how that works in practice. Do you just keep running it until the tests pass? I guess with formal verification you can run it as many times as you need, right?

ainch2• 4 months ago

pass@k means that you run the model k times and give it a pass if any of the answers is correct. I guess Lean is one of the few use cases where pass@k actually makes sense, since you can automatically validate correctness.

teekert3• 4 months ago

I also don't understand the focus on vibe coding in the marketing. Vibe coding kind of has the image of being for non-devs, right?

I do like agents (like Claude Code), but I don't consider myself to be vibe coding when I use them. Either I'm using a language/framework I know and check every step. OR I'm learning, checking every step and asking for explanations.

I tried vibe coding, and really dislike the feeling I have when doing it. It feels like building a house, but without caring about it, and just using whatever tech. Sure I may have moisture problems later, but it's a throwaway house anyway. That's how I feel about it. Maybe I have a wrong definition.

Maybe it's good to not use "vibe coding" as a synonym for programming with agent assistance. Just to protect our profession. Like: "Ah you're vibing" (because you have Claude Code open), "No, I'm using CC to essentially type faster and prevent syntax errors and get better test coverage, maybe to get some smart solutions without deep research. But I understand and vouch for every loc here. 'We are not the same.'"

DANmode• 4 months ago

It feels like building a house, but without caring about it, and just using whatever tech.

So, most homebuilders (in the US) unfortunately.

benterix• 4 months ago

I tried vibe coding, and really dislike the feeling I have when doing it. It feels like building a house, but without caring about it, and just using whatever tech. Sure I may have moisture problems later, but it's a throwaway house anyway. That's how I feel about it. Maybe I have a wrong definition.

No, I feel the same. I vibe-coded a few projects and after a few weeks I just threw them away, ultimately I felt I just wasted my time and wished I coudl get it back to do something useful.

andai6• 4 months ago

Yeah, the original meaning of Vibe Coding was "not looking at the code, just going on vibes", but a lot of people now use it to mean "AI was involved in some way".

I see a whole spectrum between those two. I typically alternate between "writing code manually and asking AI for code examples" (ChatGPT coding), and "giving AI specific instructions like, write a function blarg that does foo".

The latter I call Power Coding, in the sense of power armor, because you're still in control and mostly moving manually, but you're much stronger and faster.

I like this better than "tell agent to make a bunch of changes and come back later" because first of all it doesn't break flow (you can use a smaller model for such fine-grained changes so it goes very fast -- it's "realtime"), and second, you don't ever desync from the codebase and need to spend extra time figuring out what the AI did. Each change is sanity-checked as it comes in.

So you stay active, and the code stays slop-free.

I don't hear a lot of people doing this though? Maybe we just don't have good language for it.

lefrenchy• 4 months ago

Does Mistral come close to Opus 4.6 with any of their models?

DarkNova62• 4 months ago

Not at the moment, but a release of Mistral 4 seems close which likely bridges the gap.

re-thc• 4 months ago

Mistral Small 4 is already announced.

chucky_z• 4 months ago

I use mistral-medium-3.1 for a lot of random daily tasks, along with the vibe cli. I'd state from my personal opinion that mistral is my preferred 'model vendor' by far at this point. They're extremely consistent between releases while each of them just feels better. I also have a strong personal preference to the output.

I actively use gemini-3.1-pro-preview, claude-4.6-opus-high, and gpt-5.3-codex as well. I prefer them all for different reasons, however I usually _start_ with mistral if it's an option.

sa-code• 4 months ago

Why not Large 3? It's larger and cheaper

tjwebbnorfolk• 4 months ago

Mistral hasn't been in the running for SOTA for quite awhile now

patall2• 4 months ago

Maybe a naive question: given that they see better performance with more passes but the effect hits a limit after a few passes, would performance increase if they used different models per pass, i.e leanstral, kimi, qwen and leanstral again instead of 4x leanstral?

andai6• 4 months ago

This is called a "LLM alloy", you can even do it in agentic, where you simply swap the model on each llm invocation.

It does actually significantly boost performance. There was an article on here about it recently, I'll see if I can find it.

Edit: https://news.ycombinator.com/item?id=44630724

They found the more different the models were (the less overlap in correctly solved problems), the more it boosted the score.

patall2• 4 months ago

That sounds quite interesting. Makes me wonder if sooner or later they will have to train multiple independent models that cover those different niches. But maybe we will see that sooner or later. Thanks for the link.

jasonjmcghee5• 4 months ago

Curious if anyone else had the same reaction as me

This model is specifically trained on this task and significantly[1] underperforms opus.

Opus costs about 6x more.

Which seems... totally worth it based on the task at hand.

[1]: based on the total spread of tested models

DarkNova62• 4 months ago

I'm never sure how much faith one can put into such benchmarks but in any case the optics seem to shift once you have pass@2 and pass@3.

Still, the more interesting comparison would be against something such as Codex.

beernet• 4 months ago

Agreed. The idea is nice and honorable. At the same time, if AI has been proving one thing, it's that quality usually reigns over control and trust (except for some sensitive sectors and applications). Of course it's less capital-intense, so makes sense for a comparably little EU startup to focus on that niche. Likely won't spin the top line needle much, though, for the reasons stated.

miohtama• 4 months ago

Alignment tax directly eats to model quality, double digit percents.

hermanzegerman• 4 months ago

EU could help them very much if they would start enforcing the Laws, so that no US Company can process European data, due to the Americans not willing to budge on Cloud Act.

That would also help to reduce our dependency on American Hyperscalers, which is much needed given how untrustworthy the US is right now. (And also hostile towards Europe as their new security strategy lays out)

segmondy• 4 months ago

Ha, keep putting your prompts and workflows into cloud models. They are not okay with being a platform, they intend to cannibalize all businesses. Quality doesn't always reign over control and trust. Your data and original ideas are your edge and moat.

isodev• 4 months ago

quality usually reigns over control and trust

Most Copilot customers use Copilot because Microsoft has been able to pinky promise some level of control for their sensitive data. That's why many don't get to use Claude or Codex or Mistral directly at work and instead are forced through their lobotomised Copilot flavours.

Remember, as of yet, companies haven't been able to actually measure the value of LLMs ... so it's all in the hands of Legal to choose which models you can use based on marketing and big words.

nimchimpsky• 4 months ago

the model is open source, you can run it locally. You don't think thats significant ?

speedgoose2• 4 months ago

But you can run this model for free on a common battery powered laptop sitting on your laps without cooking your legs.

hobofan• 4 months ago

Sorry, but what are you talking about? This is a 120B-A6B model, which isn't runnable on any laptop except the most beefed up Macbooks, and then will certainly drain its battery and cook your legs.

kittikitti• 4 months ago

This is great, congratulations to the Mistral team! I'm looking forward to the code arena benchmark results. Thanks for sharing.

Havoc• 4 months ago

What are these "passes" they reference here? Haven't seen that before in LLM evals

Could definitely be interesting for having another model run over the codebase when looking for improvements

rockinghigh• 4 months ago

It's the number of attempts at answering the question.

lsb• 4 months ago

The real world success they report reminds me of Simon Willison’s Red Green TDD: https://simonwillison.net/guides/agentic-engineering-pattern...

Instead of taking a stab in the dark, Leanstral rolled up its sleeves. It successfully built test code to recreate the failing environment and diagnosed the underlying issue with definitional equality. The model correctly identified that because def creates a rigid definition requiring explicit unfolding, it was actively blocking the rw tactic from seeing the underlying structure it needed to match.

skanga• 4 months ago

TDD == Prompt Engineering, for Agentic coding tasks.

_boffin_• 4 months ago

Wild it’s taken people this long to realize this. Also lean tickets / tasks with all needed context to complete the task, including needed references / docs, places to look in source, acceptance criteria, other stuff.

jatins• 4 months ago

If Agent is writing the tests itself, does it offer better correctness guarantees than letting it write code and tests?

MillionOClock• 4 months ago

It is definitely not foolproof but IMHO, to some extent, it is easier to describe what you expect to see than to implement it so I don't find it unreasonable to think it might provide some advantages in terms of correctness.

rvz• 4 months ago

Given the issues with AWS with Kiro and Github, We already have just a few high-profile examples of what happens when AI is used at scale and even when you let it generate tests which is something you should absolutely not do.

Otherwise in some cases, you get this issue [0].

[0] https://sketch.dev/blog/our-first-outage-from-llm-written-co...

bluGill5• 4 months ago

In my experience the agent regularly breaks some current features while adding a new one - much more often than a human would. Agents too often forget about the last feature when adding the next and so will break things. Thus I find Agent generated tests important as they stop the agent from making a lot of future mistakes.

saberience• 4 months ago

That article is literally a definition of TDD that has been around for years and years. There's nothing novel there at all. It's literally test driven development.

flakiness• 4 months ago

FYI The Lean 4 paper: https://dl.acm.org/doi/10.1007/978-3-030-79876-5_37

elAhmo• 4 months ago

I don’t know a single person using Mistral models.

pelagicAustral• 4 months ago

Me neither, they're not ready for prime imo. I have a yearly sub and the product is just orders of magnitude behind Anthropic's offering. I use Code for real world stuff and I am happy with the result, Mistral is just not something I can trust right now.

consumer451• 4 months ago

Isn't their latest speech to text model SOTA? When I tested it on jargon, it was amazing.

https://news.ycombinator.com/item?id=46886735

troyvit• 4 months ago

I'm using this model for my first python project, coding using opencode along with devstral and Mistral Large 3. I know it's not as capable as other, more expensive models, but working with it this way is teaching me python. More directly to your point though, the speech to text model is really good.

It's funny because I just took a break from it to read some hn and found this post.

Adrig• 4 months ago

I used Ministral for data cleaning.

I was surprised: even tho it was the cheapest option (against other small models from Anthropic) it performed the best in my benchmarks.

Bombthecat• 4 months ago

Mistral is super smart in smaller context and asking questions about it

badsectoracula• 4 months ago

Pretty much all of my LLM usage has been using Mistral's open source models running on my PC. I do not do full agentic coding as when i tried it with Devstral Small 2 it was a bit too slow (though if i could get 2-3 times the speed of my PC from a second computer it'd be be a different story and AFAIK that is doable if i was willing to spend $2-3k on it). However i've used Mistral's models for spelling and grammar checks[0], translations[1][2], summaries[3] and trying to figure out if common email SPAM avoidance tricks are pointless in the LLM age :-P [4]. FWIW that tool you can see in the shots is a Tcl/Tk script calling a llama.cpp-based command-line utility i threw together some time ago when experimenting with llama.cpp.

I've also used Devstral Small to make a simple raytracer[5][6] (it was made using the "classic" chat by copy/pasting code, not any agentic approach and i did fix bits of it in the process) and a quick-and-dirty "games database" in Python+Flask+Sqlite for my own use (mainly a game backlog DB :-P).

I also use it to make various small snippets, have it generate some boilerplate stuff (e.g. i have an enum in C and want to write a function that prints names for each enum value or have it match a string i read from a json file with the appropriate enum value), "translate" between languages (i had it recently convert some matrix code that i had written in Pascal into C), etc.

[0] https://i.imgur.com/f4OrNI5.png

[1] https://i.imgur.com/Zac3P4t.png

[2] https://i.imgur.com/jPYYKCd.png

[3] https://i.imgur.com/WZGfCdq.png

[4] https://i.imgur.com/ytYkyQW.png

[5] https://i.imgur.com/FevOm0o.png (screenshot)

[6] https://app.filen.io/#/d/e05ae468-6741-453c-a18d-e83dcc3de92... (C code)

[7] https://i.imgur.com/BzK8JtT.png

Fnoord• 4 months ago

I use them solely.

ainch2• 4 months ago

That's likely because they're chasing enterprise - see deals with HSBC, ASML, AXA, BNP Paribas etc... Given swelling anti-US sentiment and their status as a French 'national champion', Mistral are probably in a strong position for now regardless of model performance, research quality or consumer uptake.

brainless• 4 months ago

I'm building a knowledge graph on personal data (emails, files) with Ministral 3:3b. I try with Qwen 3.5:4b as well but mostly Ministral.

Works really well. Extracts companies you have dealt with, people, topics, events, locations, financial transactions, bills, etc.

miacycle2• 4 months ago

The TDD foundation! We might need one of those. :)

JoshTriplett• 4 months ago

Pleasant surprise: someone saying "open source" and actually meaning Open Source. It looks like the weights are Apache-2.0 licensed.

jasonjmcghee5• 4 months ago

Based on community definitions I've seen, this is considered "open weights". If you can't reproduce the model, it's not "open source"

xpe3• 4 months ago

Yes “open weights” conveys the reality more clearly: merely having the parameters is very different than able to run a process that creates them. Without openness of the full process start to finish, much is hidden.*

Remember, language is what we make it. Dictionaries are useful catalogs of usage but we make the judgment calls.

* Even with the process, much is not well understood! / The ethics of releasing an open weights model at some capability level is a separate discussion.

esperent• 4 months ago

I absolutely called this a couple of weeks ago, nice to be vindicated!

I'm interested to see what it is in the age of LLMs or similar future tools. I suspect a future phase change might be towards disregarding how easy it is for humans to work with the code and instead focus on provability, testing, perhaps combined with token efficiency.
Maybe Lean combined with Rust shrunk down to something that is very compiler friendly. Imagine if you could specify what you need in high level language and instead of getting back "vibe code", you get back proven correct code, because that's the only kind of code that will successfully compile.

https://news.ycombinator.com/item?id=47192116

AlotOfReading5• 4 months ago

It's important to keep in mind that no proof system ensures your proof is the correct proof, only that it's a valid proof. Completely understanding what a proof proves is often nearly as difficult as understanding the program it's proving. Normally you benefit because the process of building a proof forces you to develop your understanding more fully.

specvsimpl• 4 months ago

Uhm, no? Even with "simple" examples like Dijkstra's shortest path, the spec is easier than the implementation. Maybe not for you, but try it out on an arbitrary 5-yr old. On the extreme end, you have results in maths, like Fermat's Last Theorem. Every teenager can understand the statement (certainly after 10 mins of explanation) but the proof is thousands of pages of super-specialized maths. It is a spectrum. For cryptography, compression, error-correction, databases, etc, the spec is often much simpler than the implementation.

hnipps• 4 months ago

Here we go.

htrp• 4 months ago

is the haiku comparison because they've distilled from the model?

rothific• 4 months ago

There have been a lot of conversations recently about how model alignment is relative and diversity of alignment is important - see the recent podcast episode between Jack Clark (co-founder of Anthropic) and Ezra Klein.

Many comments here point out that Mistral's models are not keeping up with other frontier models - this has been my personal experience as well. However, we need more diversity of model alignment techniques and companies training them - so any company taking this seriously is valuable.

nicman23• 4 months ago

they ll get there

piyh• 4 months ago

Automated theorem provers running on a $5k piece of hardware is a cool version of the future

jasonjmcghee5• 4 months ago

Curious if pass@2 was tested for haiku and sonnet?

drdaeman• 4 months ago

Can someone please explain... If I don't know any Lean (and I suspect most people don't), is it of any direct value? Trying to understand if there's something it can help me with (e.g. automatically write proofs for my Go programs somehow... I'm not sure) or should I just cheer solely for more open models out there, but this one isn't for me?

TimTheTinker2• 4 months ago

Presumably the idea is that an agent generates a Lean4 specification against which the software is measured.

But then the Lean4 specification effectively becomes the software artifact.

And we're sort of back to square 1. How do you verify a Lean4 spec is correct (and that it describes what needs to be built in the first place) without human review?

naasking3• 4 months ago

And we're sort of back to square 1.

Specifications are smaller than the full code, just as high level code is smaller than the functionally equivalent assembly. As we ascend the abstraction ladder the amount of reading a human needs to do decreases. I don't think this should really count as "back to square 1".

cadamsdotcom3• 4 months ago

It’s great to see this pattern of people realising that agents can specify the desired behavior then write code to conform to the specs.

TDD, verification, whatever your tool; verification suites of all sorts accrue over time into a very detailed repository of documentation of how things are supposed to work that, being executable, puts zero tokens in the context when the code is correct.

It’s more powerful than reams upon reams of markdown specs. That’s because it encodes details, not intent. Your intent is helpful at the leading edge of the process, but the codified result needs shoring up to prevent regression. That’s the area software engineering has always ignored because we have gotten by on letting teams hold context in their heads and docs.

As software gets more complex we need better solutions than “go ask Jim about that, bloke’s been in the code for years”.

refulgentis4• 4 months ago

I've seen this sentiment and am a big fan of it, but I was confused by the blog post, and based on your comment you might be able to help: how does Lean help me? FWIW, context is: code Dart/Flutter day to day.

I can think of some strawmen: for example, prove a state machine in Lean, then port the proven version to Dart? But I'm not familiar enough with Lean to know if that's like saying "prove moon made of cheese with JavaScript, then deploy to the US mainframe"

Paracompact3• 4 months ago

I don't think he's referring to Lean specifically, but any sort of executable testing methodology. It removes the human in the loop in the confidence assurance story, or at least greatly reduces their labor. You cannot ever get such assurance just by saying, "Well this model seems really smart to me!" At best, you would wind up with AI-Jim.

(One way Lean or Rocq could help you directly, though, would be if you coded your program in it and then compiled it to C via their built-in support for it. Such is very difficult at the moment, however, and in the industry is mostly reserved for low-level, high-consequence systems.)

baq2• 4 months ago

yesterday I had to tell a frontier model to translate my code to tla+ to find a tricky cache invalidation bug which nothing could find - gpt 5.4, gemini 3.1, opus 4.6 all failed. translation took maybe 5 mins, the bug was found in seconds, total time to fix from idea to commit - about 15 mins.

if you can get a model to quickly translate a relevant subset of your code to lean to find tricky bugs and map lean fixes back to your codebase space, you've got yourself a huge unlock. (spoiler alert: you basically can, today)

tonymet• 4 months ago

AI is the reality that TDD never before had the opportunity to live up to

nextos• 4 months ago

Not just TDD. Amazon, for instance, is heading towards something between TDD and lightweight formal methods.

They are embracing property-based specifications and testing à la Haskell's QuickCheck: https://kiro.dev

Then, already in formal methods territory, refinement types (e.g. Dafny, Liquid Haskell) are great and less complex than dependent types (e.g. Lean, Agda).

BowBun• 4 months ago

I feel like the difference is minimal, if not entirely dismissable. Code in this sense is just a representation of the same information as someone would write in an .md file. The resolution changes, and that's where both detail and context are lost.

I'm not against TDD or verification-first development, but I don't think writing that as code is the end-goal. I'll concede that there's millions of lines of tests that already exist, so we should be using those as a foundation while everything else catches up.

cadamsdotcom3• 4 months ago

Say you describe your kitchen as “I want a kitchen” - where are the knives? Where’s the stove? Answer: you abdicated control over those details, so it’s wherever the stochastic parrot decided to put them, which may or may not be where they ended up last time you pulled your LLM generate-me-a-kitchen lever. And it may not be where you want.

Don’t like the layout? Let’s reroll! Back to the generative kitchen agent for a new one! ($$$)

The big labs will gladly let you reroll until you’re happy. But software - and kitchens - should not be generated in a casino.

A finished software product - like a working kitchen - is a fractal collection of tiny details. Keeping your finished software from falling apart under its own weight means upholding as many of those details as possible.

Like a good kitchen a few differences are all that stands between software that works and software that’s hell. In software the probability that an agent will get 100% of the details right is very very small.

Details matter.

chriswarbo3• 4 months ago

Tests (and type-checkers, linters, formal specs, etc.) ground the model in reality: they show it that it got something wrong (without needing a human in the loop). It's empiricism, "nullius in verba"; the scientific approach, which lead to remarkable advances in a few hundred years; that over a thousand years of ungrounded philosophy couldn't achieve.

bluGill5• 4 months ago

That’s because it encodes details, not intent.

Be careful here - make sure you encode the right details. I've seen many cases where the tests are encoding the details of how it was implemented and not what it is intended to do. This means that you can't refactor anything because your tests are enforcing a design. (refactor is changing code without deleting tests, the trick is how can you make design changes without deleting tests - which means you have to test as much as possible at a point where changing that part of the design isn't possible anyway)

necovek3• 4 months ago

While you are right that you need to be encoding the right details, I disagree on the tests enforcing a design point.

As part of the proper testing strategy, you will have tests that cover individual behavior of a small block/function (real "unit" tests), tests that cover integration points only up to the integration itself, and a small number of end-to-end or multi-component integration tests.

Only the last category should stay mostly idempotent under refactoring, depending on the type of refactor you are doing.

Integration tests will obviously be affected when you are refactoring the interfaces between components, and unit tests will be affected when you are refactoring the components themselves. Yes, you should apply the strategy that keeps it under incremental reverse TDD approach (do the refactor and keep the old interface, potentially by calling into new API from the old; then in second step replace use of old API as well, including in tests).

Tests generally define behavior and implementation in a TDD approach: it'd be weird if they do not need changing at all when you are changing the implementation.

strujillo2• 4 months ago

That matches what I’ve seen as well — generation is the easy part, validation is the bottleneck.

I’ve been experimenting with a small sparse-regression system that infers governing equations from raw data, and it can produce a lot of plausible candidates quickly. The hard part is filtering out the ones that look right but violate underlying constraints.

For example, it recovered the Sun’s rotation (~25.1 days vs 27 actual) from solar wind data, but most candidate equations were subtly wrong until you enforced consistency checks.

Feels like systems that treat verification as the source of truth (not just an afterthought) are the ones that will actually scale.

igravious• 4 months ago

"and continues to scale linearly"

it clearly and demonstrably does not. in fact, from eyeballing their chart Qwen, Kimi, and GLM scale linearly whereas Leanstral does not. But this is not surprising because the Alibaba, Moonshot, and Zhipu have hundreds of employees each and hundreds of millions of dollars of investment each.

jiehong• 4 months ago

Congratulations on the launch!

Mistral seems to focus on a different market than the others. Their best model is meh, their best ASR model locally is either rather slow compared to Parakeet on similar languages, or not as good for others (like qwen ASR).

Side note: Lean seems quite unreadable with tons of single letter variable names. Part of it is me being unaccustomed with it, but still.

aimanbenbaha• 4 months ago

Mistral seems to focus on some niche LLM model tooling that are somehow very needed in certain cases. Can't forget their OCR multimodal embedding model!

toastal• 4 months ago

Naturally the Microsoft-owned language is getting the AI hype instead of the more mature options that could do this sort of work… Agda, ATS, Coq/Rocq, Dafny, Fstar, Idris, Isabelle, Why3 just to name a few.

mrklol• 4 months ago

Am I missing something? Isn’t that the language most are using currently when looking at research at openai, google, deepseek etc?

Paracompact3• 4 months ago

A bit uncharitable. I'm a diehard fan of Rocq, but it's nothing unusual to see the young new hotness that is Lean continue to get the spotlight. It's not a sign of Microsoft putting its thumb on the scales, and the hype for Lean has long predated LLMs.

It's certainly less mature when it comes to verified programming, but its appeal to mathematicians (rather than formal methods experts) has earned it much respect.

markusde• 4 months ago

You should check out the recent PR's to the Agda repo... the community is currently very divided about AI. For better or worse, the people driving the Lean project have been interested in AI for quite some time.

westurner• 4 months ago

From https://mistral.ai/news/leanstral :

  Model        Cost ($) Score
  ..
  Claude Opus     1,650 39.6
  ..
  Leanstral pass@8  145 31.0
  Leanstral pass@16 290 31.9

wazHFsRy• 4 months ago

Is anyone using this approach with lean to ship production code? Writing lean spec as human, implementation and proof by agent? And then shipping lean or exporting to C? Would be great to understand how you are actually using this.

maelito• 4 months ago

I don't understand how this can impact my JS (+yaml, css, etc) code writing in a complex app.

blueTiger33• 4 months ago

I read it as Lanestra, and thought of that story :D

kimsant2• 4 months ago

AI agents will become a comodity.

Europeans not wanting to be dependent, and they are giving for free what US investors planed to charge with 90% margin.

Amazing! What a blast. Thank you for your service (this first 100M$ burned to POC GPT1 and from here, we are so good to go)

bigfudge• 4 months ago

I really hope you're right. Sadly, though, I don't see any evidence of UK companies disinvesting from big US tech. There aren't good alternatives and what there is is too complex. As long as 'everyone else is still using MS', it seems like it's a brave CTO that switches to European providers. Unless that happens, the network effect of having AI+data is likely to mean US tech still has a big advantage in corp settings. But, HN - please tell me I'm wrong!

worldsayshi• 4 months ago

I wonder what the biggest (non-AI) moats are for US tech against the alternatives?

utopiah• 4 months ago

There aren't good alternatives and what there is is too complex.

Sounds like a worth challenge for this community, mind giving actual examples and see what others can suggest?

baq2• 4 months ago

they will, but the jagged frontier is fractal and each one will have different capabilities; you'll want to mix models and to get best results consistently you'll need to.

warpspin2• 4 months ago

The problem with the European independence story is, that it seems Mistral runs its own stuff also on US cloud act affected infrastructure. This makes them a very weird value proposition: If I accept a level of "independence" whereby I run on AWS or Azure, I could as well pay for Anthropic or GPT to have SOTA performance.

If I do not accept that level of independence but want more, I need to buy what's on OVH, Scaleway, Ionos etc. or host my own, but that usually means even smaller, worse models or a lot of investment.

Nevertheless, the "band" that Mistral occupies for economic success is very narrow. Basically just people who need independence "on paper" but not really. Because if I'm searching for actual independence, there's no way I could give them money at the moment for one of their products and it making sense, cause none of their plans are an actual independence-improvement over, let's say, Amazon Bedrock.

I really really want to support them, but it must make economic sense for my company, too, and it doesn't.

kimsant2• 4 months ago

I don’t care about the servers, they are a comodity already.

The key is to avoid chantage, remember Oracle with DBs, people learned not to build on top of unreplaceable stuff

tin7in2• 4 months ago

They are building their own infra - south of Paris and another one was announced in Sweden recently.

atmosx• 4 months ago

lol, why does the paper abstract assume I know what Lean is and it goes on to talk about lean 4 improvements?

cicko• 4 months ago

Why do you expect to understand an article you randomly read off the interwebs?

ucsandman• 4 months ago

love the opensource push for agents, the fleet grows!

agentultra• 4 months ago

Very cool but I haven’t been able to convince software developers in industry to write property based tests. I sometimes joke that we will start writing formal proofs until the tests improve. Just so that they will appreciate the difference a little more.

I can’t even convince most developers to use model checkers. Far more informal than a full proof in Lean. Still highly useful in many engineering tasks. People prefer boxes and arrows and waving their hands.

Anyway, I don’t know that I’d want to have a system vibe code a proof. These types of proofs, I suspect, aren’t going to be generated to be readable, elegant, and be well understood by people. Like programs they generate it will look plausible.

And besides, you will still need a human to review the proof and make sure it’s specifying the right things. This doesn’t solve that requirement.

Although I have thought that it would be useful to have a system that could prove trivial lemmas in the proof. That would be very neat.

rowanG0772• 4 months ago

The point is you just need to scrutinize the theorem. Not easy either, but still significantly less work than writing the proof.

xpe3• 4 months ago

Public service announcement to hopefully reduce unnecessary knife fights*:

There are two compatible and important (but different) questions in play:

1. Is a program correct relative to a formal specification?

2. Is the formal specification what we mean/want?

*: Worth asking: “What that other person necessarily wrong? Or perhaps they are discussing a different aspect or framing?” AKA: “be curious and charitable” I’m not going to link to the specific threads, but they are happened / are happening. Le Sigh.

techcam• 4 months ago

The tricky part is that prompts can look “correct” but still behave unpredictably depending on phrasing.

strujillo2• 4 months ago

Formal verification and code synthesis feel like natural companions for automated scientific discovery. I’ve been working on a small (~800‑line) Python agent that uses sparse regression to uncover governing equations directly from data; it’s managed to validate twelve physical laws, including deriving the Sun’s rotation rate from NASA plasma measurements and correcting Gemini’s plasma conservation. Having an agent like Leanstral that can reason about proofs and specifications would be a powerful complement to data‑driven model discovery — it closes the loop between experimentation and provable correctness.

storus• 4 months ago

I just feel like Mistral is heading for bad financial times when they are focusing on fringe academic areas and not on building a business out of their research. Initial Mistral was largely based on LLaMA, then they added innovative MoE and since then disappeared, doing AI consulting for big EU companies instead.

robertwer• 4 months ago

I’ve never worked with formal validation (barely remember my CS course). This release looks impressive. But I'm trying to wrap my head around the near-term practical applications for everyday software.

Right now, we see a lot of business experts in enterprises tempted to use AI to impl. business logic so they don't have to wait for (or pay) software experts. Would this kind of technology help these users any time soon?

My current theory is that the real breakthrough for these non-developers will only happen when they can actually verify the result themselves without needing an another expert in the loop. But I don't see that with formal validation anytime soon.

Do I overlook something?

bb-connor• 4 months ago

this is very exciting work

rafph• 4 months ago

This is a typical AI announcement. Putting FLTEval scores ahead of explanations, copying code from Rocq and basically not explaining at all what the setup does.

The average quality of an AI announcement is that of a Memecoin. Lots of graphs, meandering text and no substance.

Andrei_dev• 4 months ago

Yeah, this tracks. Developers who actually read what the AI spits out catch the obvious mistakes. The ones who just tab-complete their way through a whole project don't. And where it bites you isn't where you'd expect — logic bugs get caught fast. It's the boring security stuff. No input validation, CORS wide open, admin routes with no auth at all. Formal verification tells you whether a function matches its spec. The problem with AI-generated code goes a level below that. It's everything nobody bothered specifying — like "maybe don't hardcode your database credentials."

michaelgdwn• 4 months ago

The formal verification angle is what makes this interesting. Most coding agents optimize for "code that compiles and passes tests" — that's a low bar. Curious whether the proof artifacts are persisted for audit trails or thrown away after verification.

casey2• 4 months ago

I get why the massive roll out is happening, but it's still possible that everything LLMs are good for can be done on the cheap. It's just another point for Google's multimodal focus.

I think in 10 years most providers will implode because they can't justify the debt for a cheap commodity product. While Google (and probably OpenAI) will have a huge moat due to users/multimodal/world models

Leanstral: Open-source agent for trustworthy coding and formal proof engineering

Related Stories