Dashu aims to become Rust’s full big-number stack
Dashu’s July 12 spotlight framed it as a Rust-native alternative to GMP, MPFR and MPC, with integers, floats, rationals and complex numbers in one stack.
A Chinese Rust community post on July 12 put Dashu in a different category from the usual big-number crate launch: it described a Rust-native attempt to cover the same territory as GMP, MPFR and MPC, while staying pure Rust and fully no_std-compatible. The project now spans arbitrary-precision integers, rationals, binary and decimal floating point, and complex numbers through dashu-cmplx, which makes the “full stack” claim more than branding. Even the name carries a local joke, a Chinese pun on big numbers and big trees, which fits a project that looks built for a long climb rather than a quick demo.
That breadth is the point. Dashu’s README says the crate set is meant to be a Rust-native alternative to GNU GMP plus MPFR, with MPC support planned, and that it focuses on ergonomics, readability, speed and memory usage. The code is split into small pieces, including dashu-int, dashu-float, dashu-ratio, dashu-cmplx and dashu-macros, so downstream code can pull in only the numeric layer it needs. For embedded work, no_std support removes a major blocker. For scientific and finance code, the appeal is simpler builds and one language boundary instead of a mixed C and Rust dependency chain.
The repository itself shows a project that is being actively pushed forward, not just discussed. cmpute/dashu currently lists 111 stars, 17 forks and 454 commits, and its recent history includes a Release v0.5 commit, work on complex numbers, hyperbolic functions and a Chinese README and mdBook mirror update. The current minimum supported Rust version is 1.68, and the project also ships dashu-python, which exposes integers, rationals, floats and complex numbers to Python users without requiring a Rust toolchain. A roadmap issue set two clear prerequisites for v1.0: complex number support and enough users testing the library.
That matters because Dashu is trying to replace a stack with a very deep history. GMP is a portable C library for integers, rationals and floating-point arithmetic, and it leans on CPU-specific assembly to squeeze out speed. MPFR adds multiple-precision floating-point arithmetic with correct rounding, and GNU MPC extends the same model to arbitrary-precision complex numbers. Rug already gives Rust developers a high-level bridge into that ecosystem. Dashu’s bet is that many teams would rather have the portability, package simplicity and single-language ergonomics of a native Rust stack, even if the old C path still holds the edge for the most demanding correctness and performance cases.
That is why the July 12 spotlight landed: Dashu is no longer just another precision crate competing for attention, but a credible answer to Rust’s long dependence on C numeric libraries.
This article was produced by Prism’s automated news system from verified source data, official records, and press releases, then run through automated quality and moderation checks before publishing. The system is built and supervised by the people who set the standards it runs under. Read our full AI policy.
Did this article answer your question?
