GX
BETA — Actively developed

Fast. Typed.
Native.

A statically-typed language for games, systems, and graphics. No build system needed — just write code and compile. Two native backends — C transpiler and LLVM IR — both shipping. GPU compute and AI support on the roadmap.

v0.8.3-beta — Windows, macOS, Linux
game.gx
struct Player {
name: str
health: i32
pos: vec3
}

fn main() {
var p = Player{"Alice", 100, vec3{0.0, 1.0, 0.0}}
p.health -= 10
print("Player {p.name} at ({p.pos.x}, {p.pos.y}, {p.pos.z})")
}
.gx Lexer Parser Type Check Backend
C Backend Default — compiles everywhere C does
LLVM Backend --backend llvm — LLVM IR via clang
JS Backend Web & WASM — powers the Playground
GPU Compute Planned — compute shaders from GX

Why GX?

A language designed for people who want modern syntax without giving up control over performance and memory.

C Performance, Modern Syntax

No garbage collector, no runtime. GX compiles to plain C, then to native code.

🛠

Two Native Backends

Ship through the bundled C transpiler for fast incremental builds, or switch to --backend llvm for LLVM-based codegen via clang. Same language, same modules, same output.

Built-in Graphics Types

Native vec2/3/4 and mat2/3/4 with operator overloading and swizzle (v.xy, v.zyx).

🔗

First-Class C Interop

Use any C library with @c_include and extern declarations. No bindings generator needed.

Compile-Time System

#if, #for, #fn, #type — conditional compilation, loop unrolling, compile-time functions, and monomorphized templates. All resolved before the C compiler sees the code.

📦

Self-Contained Builds

@link("opengl32") in your source file — no CMake, no Makefile.

Compile-Time Reflection

@fields(T), @field(T, i), @members(E) — iterate over struct fields and enum members at compile time.

On the Roadmap

GX is in beta and actively developed. Here is what is coming next.

Planned

Interpreter Backend

A direct-execution backend for fast iteration and scripting — no C compiler round-trip. Complements the native C and LLVM backends.

🎮 Planned

GPU Compute

Run parts of your program on the GPU as compute shaders. No separate shader language, no manual dispatch. Just mark the code and go.

🤖 Planned

AI Support

Tensor types and GPU execution for machine learning workloads. Train and infer directly in GX without external frameworks.

Planned

LSP & More IDEs

Language Server Protocol for VS Code, Sublime, and other editors. IntelliJ plugin already available.

Quick Taste

Built-in math, string interpolation, and compile-time reflection.

math_demo.gx
import math
struct Vec { x:f32 y:f32 z:f32 }


fn main() {
var angle: f32 = radians(45.0)
var result = sinf(angle)
print("sin(45) = {result}\n")

print("Vec has {@fields(Vec)} fields")
}

Ready to try GX?

Get the compiler, read the docs, or try it in your browser.

Get notified when new builds drop

GX is in active beta. Drop your email and we’ll ping you when the next compiler build, module, or editor plugin lands. No spam, just release notes.

Which platforms do you build for? (pick at least one)

Want the longer pitch? /beta