No CMS or framework detected, hosted on Cloudflare. Domain registered in 2025.
logicaffeine.com is hosted on Cloudflare.
| Technology | Version | First seen |
|---|---|---|
| CDN & Hosting1 | ||
| – | 7 Nov 2026 | |
| Compiled-to-Web & WebAssembly3 | ||
| – | 7 Nov 2026 | |
| Fonts & Icon Fonts1 | ||
| – | 7 Nov 2026 | |
Registration from the registry's RDAP record, the certificate the site serves, and its rank in public traffic lists.
The crew hasn’t read this site’s colors yet. We read a palette when we capture a site’s screenshot, busiest sites first. Meanwhile, browse palettes from other sites.
novalee.ns.cloudflare.comrenan.ns.cloudflare.comgoogle-site-verification=IPkJeFyuGSEGM23T-n4lnYJstI0yZrzMBNaxFd2G5bkUser-agent: * Allow: / Sitemap: https://logicaffeine.com/sitemap.xml
# Logicaffeine > LOGOS is an English programming language. Write English sentences and get First-Order Logic; write English programs and get running code, a web app, or Rust, C, WASM and RISC-V. Current release: 0.11.0. ## What it is Logicaffeine is a natural-language compiler with two modes that share one parser. - **Logic mode** turns English into First-Order Logic (Unicode, LaTeX, ASCII or Kripke form) with neo-Davidsonian event semantics, modal and temporal logic, and every reading of an ambiguous sentence: "Every woman loves a man." comes back with both quantifier scopes. - **Imperative mode** compiles LOGOS programs: static types, ownership, generics, pattern matching, contracts, modules, tests written in the language (`## Test`, `Expect that …`, `For all` properties), structured concurrency on a seeded, replayable scheduler, and CRDTs for shared state. - **LGX**, the web framework: a page is one file (`State:`, `Actions:` in English, `View:` HTML, `Style:` CSS) compiled to WebAssembly, and `## Route`s ending in `Serve on port 8080.` are the server. - **Execution:** a tree-walking interpreter, a register bytecode VM, EXODIA (a copy-and-patch JIT), and ahead-of-time Rust, C, direct WASM and RISC-V backends; `largo emit` writes CUDA, Tenstorrent and other device bundles. - **Proofs:** a Calculus of Constructions kernel with an independent second checker, SAT solving with DRAT/LRAT certificates, Lean 4 import re-checked by both kernels, and optional Z3 static verification. A model asked to formalize a sentence guesses one reading; LOGOS parses, so it returns them all, and a proof counts only when the kernel checks it. ## Get started - Install (Linux, macOS): `curl -fsSL https://logicaffeine.com/install.sh | sh` - Install (Windows, in PowerShell): `irm https://logicaffeine.com/install.ps1 | iex` - Pairing with a person, step by step (install, learn the language, start a project, then code, logic, math and hardware): https://logicaffeine.com/agent.txt - First program, no Rust toolchain needed: `largo new hello && cd hello && largo run --interpret` - Getting started guide: https://github.com/Brahmastra-Labs/logicaffeine/blob/main/docs/getting-started.md - In the browser, no install: https://logicaffeine.com/studio ## Documentation - Guides: https://github.com/Brahmastra-Labs/logicaffeine/blob/main/docs/README.md - Every construct on one page: https://github.com/Brahmastra-Labs/logicaffeine/blob/main/LOGOS_QUICKGUIDE.md - Rust API documentation: https://docs.logicaffeine.com - Everything in one file for language models: /llms-full.txt - Machine-readable companions: /errors.json (every diagnostic code with its Socratic explanation) and /examples.json (the runtime-verified example corpus with expected outputs) - The toolchain is an MCP server: `claude mcp add logos -- largo mcp` ## Main Pages ### / (Homepage) "Code we can all read. Logic and mathematics built in. Bring your own agent." The agent instructions a person pastes into their coding agent (install largo, read /llms-full.txt, start a project, then code, logic, math and hardware with the released largo), copied in one click. Below them, a mini Studio: a LOGOS web page in the Studio's editor beside its live preview, and a Logic, a Math and a Hardware example to run. ### /studio Interactive three-panel IDE with Logic, Code, Math, Hardware and Web modes: - File browser with example files - Live editor with syntax highlighting - Context-sensitive output (FOL formulas, AST trees, REPL, proofs) Free to use, no account required. ### /learn A roadmap for learning First-Order Logic, one lesson at a time: - Era 1: First Steps (Introduction, Syllogistic Logic, Fallacies) - Era 2: Building Blocks (Propositional Logic, Proofs) - Era 3: Expanding Horizons (Quantificational, Modal, Deontic, Belief Logic) - Era 4: Mastery (Ethics, Metalogic, History, Deviant Logic, Philosophy) Each lesson gives two examples and one question, with a hint after a wrong answer and optional extra practice. ### /guide The interactive syntax guide; every example runs in place: - LOGOS syntax and grammar - Lexicon configuration - First-Order Logic concepts - Code examples with Run/Copy/Reset ### /contact Contact the team. LOGOS is open source under the MIT License. The page holds the install commands, a form that reaches the team, and the FAQ. ### /crates The crates published to crates.io, each described in its own manifest's words: - logicaffeine-base: Foundation types for Logicaffeine: arenas, string interning, spans, and exact integers, decimals, quantities, money, dates and UUIDs - logicaffeine-cli: largo — the LOGOS build tool: projects, builds, tests, serving, proofs, and the package registry - logicaffeine-compile: The LOGOS compiler: type checking, the optimizer, the interpreter and bytecode VM, and the Rust, C, WASM, RISC-V, CUDA and hardware backends - logicaffeine-data: LOGOS runtime values, collections, wire format and CRDTs, shared by native and WASM builds - logicaffeine-forge: The machine-code layer of the LOGOS copy-and-patch JIT: W^X executable memory, build-time stencils, a register-allocating x86-64 backend - logicaffeine-jit: Native execution tier for the LOGOS bytecode VM: hot functions and loops tier up to x86-64 machine code - logicaffeine-kernel: Calculus of Constructions proof kernel with inductive types, universes, quotients, a second checker and Lean 4 import - logicaffeine-language: The LOGOS front end: English sentences to first-order logic, and the parser for LOGOS programs and LGX pages - logicaffeine-lexicon: The English vocabulary behind LOGOS: verb, noun and adjective metadata for build-time lookup tables, plus an optional runtime JSON index - logicaffeine-lsp: The LOGOS language server (LSP): diagnostics, completion, hover, navigation and refactoring for .lg files, literate .md files and LGX pages - logicaffeine-proof: Kernel-checked proof search and tactics, certified SAT/CDCL with DRAT/LRAT output, and a number-theory and cryptanalysis library - logicaffeine-runtime: Deterministic, replayable concurrency runtime for the LOGOS interpreter and VM: task scheduler, channels, Select, timers, seed and trace - logicaffeine-system: Platform IO, networking, HTTP serving, persistence, and post-quantum crypto (Keccak/SHA-3, ChaCha20-Poly1305, ML-KEM, ML-DSA) for LOGOS - logicaffeine-tv: Translation validation for LOGOS: a symbolic encoder of the source, checked with Z3 against the tree-walking interpreter per program - logicaffeine-verify: Z3-based static verification and model checking for Logicaffeine ### /roadmap Development roadmap with 13 milestones: 1. Core Transpiler (Complete) 2. Web Platform (Complete) 3. Imperative Language (Complete) 4. Type System (Complete) 5. Concurrency & Actors (Complete) 6. Distributed Systems (Complete) 7. Security & Policies (Complete) 8. Proof Assistant (Complete) 9. Native Compilation Tier (Complete) 10. Hardware Verification (Complete) 11. Translation Validation (In Progress) 12. Quantum Backend (Planned) 13. Silicon (In Progress) ### /benchmarks Live benchmark results with full methodology: 35 programs across 11 languages (C, C++, Rust, Zig, Go, Java, JavaScript, Python, Ruby, Nim, LOGOS), the certified-solver-vs-the-field comparison (Z3, Kissat, SaDiCaL, CaDiCaL, CryptoMiniSat), and the wire-codec head-to-head (Protobuf, Cap'n Proto, MessagePack, bincode, postcard, CBOR, Arrow, JSON). Every number links to raw JSON and reproducible harnesses. ### /registry Package registry for community-contributed logic modules. Search, browse, and publish packages with GitHub authentication. ### /news Release notes, tutorials, and announcements — one article per release with engineering deep-dives (each article is statically served at /news/<slug>; the full list is in /sitemap.xml). ### /privacy Privacy policy. ### /terms Terms of service. ## Technical Details ### Supported Logic Systems - Classical propositional logic - First-order predicate logic - Modal logic (necessity, possibility) - Deontic logic (obligation, permission) - Epistemic logic (belief, knowledge) - Temporal logic (past, future, always) ### Architecture - **Frontend**: Dioxus 0.7 (Rust) compiled to WebAssembly; content pages are statically prerendered for crawlers - **Backend**: Rust with arena-allocated AST, symbol interning - **Logic pipeline**: Input → Lexer → Parser → AST → Transpiler → FOL Output - **Imperative pipeline**: five execution tiers — tree-walking interpreter, register bytecode VM, copy-and-patch JIT, AOT-to-Rust, and a direct WASM backend - **Proof Engine**: pure-Rust certified prover (SAT/CDCL, GF(2)/GF(p) algebraic routes, SR symmetry proofs) checked by a CoC proof kernel; Z3 powers the optional native static-verification tier ### Performance Measured by the open benchmark suite (35 programs x 11 languages, results at /benchmarks): - LOGOS release binaries: 4.16x geometric-mean speedup vs C — faster than Rust (1.05x), Zig (1.09x), and C++ (0.99x) on the same suite - The certified solver beats Z3, Kissat, CaDiCaL, and CryptoMiniSat by orders of magnitude on structured UNSAT families (pigeonhole, Tseitin parity, mod-p counting) - The wire codec produces the smallest encoding on every fair workload vs Protobuf, Cap'n Proto, MessagePack, bincode, postcard, and CBOR ### Lexicon System Vocabulary database (lexicon.json) defining: - Verbs with Vendler class and transitivity - Nouns with animacy, gender, and number - Adjectives with intersective/subsective properties - 2000+ vocabulary entries ## Technical Details ### Supported Logic Systems - Classical propositional logic - First-order predicate logic - Modal logic (necessity, possibility) - Deontic logic (obligation, permission) - Epistemic logic (belief, knowledge) - Temporal logic (past, future, always) ### Architecture - **Frontend**: Dioxus 0.7 (Rust) compiled to WebAssembly; content pages are statically prerendered for crawlers - **Engine**: Rust, with an arena-allocated AST and symbol interning; the same engine runs natively and in the browser as WebAssembly - **Logic pipeline**: Input → Lexer → Parser → AST → Transpiler → FOL Output - **Imperative pipeline**: a tree-walking interpreter, a register bytecode VM, a copy-and-patch JIT, and ahead-of-time Rust, C, direct WASM and RISC-V backends - **Proof Engine**: pure-Rust certified prover (SAT/CDCL, GF(2)/GF(p) algebraic routes, SR symmetry proofs) checked by a CoC proof kernel; Z3 powers the optional native static-verification tier ### Performance Measured by the open benchmark suite (35 programs x 11 languages, results at /benchmarks): - LOGOS release binaries: 4.16x geometric-mean speedup vs C — faster than Rust (1.05x), Zig (1.09x), and C++ (0.99x) on the same suite - The certified solver beats Z3, Kissat, CaDiCaL, and CryptoMiniSat by orders of magnitude on structured UNSAT families (pigeonhole, Tseitin parity, mod-p counting) - The wire codec produces the smallest encoding on every fair workload vs Protobuf, Cap'n Proto, MessagePack, bincode, postcard, and CBOR ### Lexicon System Vocabulary database (lexicon.json) defining: - Verbs with Vendler class and transitivity - Nouns with animacy, gender, and number - Adjectives with intersective/subsective properties ## Links - Website: https://logicaffeine.com - GitHub: https://github.com/Brahmastra-Labs/logicaffeine - Getting started: https://github.com/Brahmastra-Labs/logicaffeine/blob/main/docs/getting-started.md - Guides: https://github.com/Brahmastra-Labs/logicaffeine/blob/main/docs/README.md - Rust API documentation: https://docs.logicaffeine.com - Syntax guide: https://logicaffeine.com/guide - Crates: https://logicaffeine.com/crates - Studio: https://logicaffeine.com/studio - Learn: https://logicaffeine.com/learn - Benchmarks: https://logicaffeine.com/benchmarks - News: https://logicaffeine.com/news - Changelog: https://github.com/Brahmastra-Labs/logicaffeine/blob/main/CHANGELOG.md - X (Twitter): https://x.com/logicaffeine - All indexable pages: https://logicaffeine.com/sitemap.xml ## Contact - Email: team@datacenter.dev - Reach the team: https://logicaffeine.com/contact#contact - Discord: https://discord.gg/logicaffeine ## Full documentation in one file The quick guide, every guide and the error catalog, as one file for language models: /llms-full.txt. ## Organization Logicaffeine is developed by Datacenter.Dev. ## License MIT License. The whole of it is LICENSE.md in the repository. --- This file follows the llms.txt convention for language models and AI search. People should start at https://github.com/Brahmastra-Labs/logicaffeine/blob/main/docs/getting-started.md
| – |
| 7 Nov 2026 |
| – | 7 Nov 2026 |