
Gossamer 0.1.0 released with Rust syntax and Swift-like memory model
Gossamer 0.1.0 offers Rust-style syntax, goroutine-like concurrency, and a Swift-inspired ARC memory model, compiling to LLVM for Linux, macOS, and Windows [Gossamer site].
Gossamer 0.1.0 was released on 2026-07-03, delivering a compiler that translates Rust-flavored syntax into LLVM IR and ships with a standard library that implements lightweight goroutine-style threads and a Swift-like automatic reference counting (ARC) memory model [Gossamer site]. The language's documentation lists a "Hello, world" program that compiles in under 200 ms on a 2024-generation Intel i7 [Gossamer site].
The initial release includes Rust-inspired syntax with pattern matching, ownership annotations, and a Cargo-compatible package manager. Gossamer also features real goroutines, scheduling millions of concurrent tasks with a single OS thread, exposing spawn and await primitives that map directly to the underlying LLVM scheduler [Lobsters]. Additionally, it has a Swift-style ARC, where objects are deallocated deterministically when the last strong reference drops, eliminating the need for manual drop calls while preserving Rust’s safety guarantees.
Gossamer's cross-platform LLVM backend generates binaries for x86_64-linux, aarch64-macOS, and x86_64-Windows out of the box, and the ffi module lets developers call C functions without unsafe wrappers. The compiler itself is written in Rust, keeping the toolchain self-hosted [Gossamer site].
Gossamer's hybrid memory model combines Rust's ownership model with ARC, allowing engineers to write low-level services that free resources exactly when they become unreachable, reducing latency spikes common in GC-based runtimes [Gossamer site]. The goroutine runtime also enables network servers to spawn thousands of connections with a single line of code, a pattern demonstrated in the language's benchmark suite (1 M concurrent echo connections on a single core) [Lobsters]. Furthermore, targeting LLVM means existing tooling—profilers, debuggers, and static analysers—works unchanged, lowering the barrier for teams adopting a new language without rewriting their DevOps pipelines [Gossamer site].
Subscribe to the broadcast.
Daily digest of the day's most important tech news. No fluff. Engineering signal only.
// delivered via substack · double-opt-in confirmation


