SaaSHub helps you find the best software and product alternatives Learn more →
Top 23 Lean lean4 Projects
-
Project mention: Anthropic lanza Fable 5.2 y Mythos 5.2 a $8 por millón de tokens de entrada | dev.to | 2026-11-09
-
SaaSHub
SaaSHub - Software Alternatives and Reviews. SaaSHub helps you find the best software and product alternatives
-
Project mention: AI Agents Need More Than Better Prompts: They Need a Physics Framework | dev.to | 2026-10-18
Lean is an interactive theorem prover and programming language. Its kernel checks proof terms against formal propositions and their definitions and assumptions. Mathlib supplies a large reusable library of formal mathematics. These are powerful tools for asking whether a precise mathematical statement follows from declared foundations.
-
Project mention: Where machine-checked mathematics pays: prizes, bounties and registries for Lean proofs (October 2026) | dev.to | 2026-10-13
Formal Conjectures (https://github.com/google-deepmind/formal-conjectures) wants Lean statements of open conjectures, fixes of misformalisations, and open-to-solved updates; proofs over 25-50 lines are hosted elsewhere and linked. Follows Mathlib's AI policy; needs a Google CLA.
-
-
batteries
The "batteries included" extended library for the Lean programming language and theorem prover (by leanprover-community)
-
-
-
-
TorchLean
TorchLean is the first unified Lean 4 framework for neural-network specification, execution, and verification.
Project mention: TorchLean is a Lean 4 framework for neural-networks | news.ycombinator.com | 2026-10-28 -
sparkle
A type-safe, formally verifiable HDL compiler in Lean 4. Inspired by Clash, built for high-assurance hardware synthesis. (by Verilean)
-
-
verified-3d-mesh-intersection
Formally verified 3D mesh intersection - trust 93 lines of spec, not 1000+ lines of AI-written code
Project mention: Show HN: Formally verified 3D CSG: Trust 93 lines spec, not 1000 lines AI code | news.ycombinator.com | 2026-07-28and bugs in an alternative C++ implementation that also used exact rationals mentioned in https://github.com/schildep/verified-3d-mesh-intersection#co....
Also note that many implementations don’t even try to treat special geometrical cases correctly (for example they „cheat“ by using something called simulation of simplicity which provides some guarantees against some bugs, but at the expense of not treating the cases exactly correctly which can lead to artifacts).
-
-
-
I explicitly give credit to the papers the proof builds upon. Quoting from the article:
>The proof was only possible thanks to the many existing results from [References](https://gaearon.github.io/conway-refinement/#/references). In particular, [A factorisation theory for generalised power series and omnific integers](https://doi.org/10.1016/j.aim.2024.109513) by S. L’Innocente and V. Mantova has played a crucial role in the proof.
The project README (https://github.com/gaearon/conway-refinement) also states:
>Over the years, Berarducci, Pitteloud, Pommersheim and Shahriari, and L'Innocente and Mantova developed increasingly strong results about exactly that. This proof builds on all of that work and claims to finish the refinement conjecture.
It includes the same list of references: https://github.com/gaearon/conway-refinement#references
Regarding who I've talked to over email, I wasn't sure I want to "out" him explicitly in the article due to the article's likely polarizing nature, but it is Prof. Mantova (a co-author of the primary paper that the proof I published builds on). You can read his reflection here: https://news.ycombinator.com/item?id=49761718
I have not had an extended conversation with anyone else.
-
SafeVerify
A Lean4 script for robustly verifying submitted proofs of theorems and implementations of functions
Project mention: AlphaProof Nexus Solved 9 Open Erdős Problems. The Lesson for AI Engineers Is the Verifier, Not the Model | dev.to | 2026-10-20SafeVerify (proof validator used by the agents): https://github.com/GasStationManager/SafeVerify
-
Project mention: Mathematical finance, formally verified in Lean 4 | news.ycombinator.com | 2026-06-29
-
hesper
Verified GPU programming framework for Lean 4. Write type-safe WebGPU shaders with formal verification, hardware-accelerated matrix ops, and cross-platform support (Metal/Vulkan/D3D12). Build provably correct GPU compute and ML inference engines.
Some may also find Junji Hashimoto's GPU programming library in lean (w/ webgpu) interesting:
https://github.com/Verilean/hesper
Even includes an example of transformer inference (quantized 1.5 bit):
https://github.com/Verilean/hesper/blob/a688ce9848d6416b2e95...
-
I don't use tla+ to model real-world systems anymore, Claude is able to model systems in Lean 4 and the binary executable can handle real input or I can directly generate c on numeric types that have ring structure (integers, rationals, bits).
https://github.com/lambdaclass/truth_research_zk?ref=blog.la...
-
-
-
leandoom
(Lean)DOOM: a native Lean 4 DOOM engine with original-WAD support and kernel-checked movement and autoaim proofs.
Lean lean4 discussion
Lean lean4 related posts
-
Anthropic lanza Fable 5.2 y Mythos 5.2 a $8 por millón de tokens de entrada
-
Show HN: LTC Bounty to find a bug in Lean 4 formal Yang-Mills proof
-
OpenAI Math Proofs Explained: Has AI Solved Math Problems? What the 722 Manuscripts Really Show
-
Writing about pratt parsing so I remember it
-
Working with academic communities to build scientific understanding
-
Navier–Stokes Lost in Translation
-
Rails World 2026 Opening Keynote [video]
-
A note from our sponsor - SaaSHub
www.saashub.com | 11 Nov 2026
Index
What are some of the best open-source lean4 projects in Lean? This list will help you:
| # | Project | Stars |
|---|---|---|
| 1 | lean4 | 9,915 |
| 2 | mathlib4 | 4,442 |
| 3 | formal-conjectures | 1,353 |
| 4 | SciLean | 544 |
| 5 | batteries | 432 |
| 6 | NNG4 | 389 |
| 7 | lean4-metaprogramming-book | 320 |
| 8 | lean-smt | 315 |
| 9 | ProofWidgets4 | 232 |
| 10 | TorchLean | 208 |
| 11 | sparkle | 169 |
| 12 | lean4-raytracer | 146 |
| 13 | verified-3d-mesh-intersection | 110 |
| 14 | con-nf | 89 |
| 15 | lean4-maze | 72 |
| 16 | conway-refinement | 56 |
| 17 | SafeVerify | 51 |
| 18 | formal-mathfin | 51 |
| 19 | hesper | 50 |
| 20 | truth_research_zk | 34 |
| 21 | interval | 29 |
| 22 | lean4-analysis-tao | 27 |
| 23 | leandoom | 21 |