Documents Cargo build targets, generated artifact placement, Visual Studio NMake wrappers, PowerShell entry points, and reproducible release requirements.
Implementation evidence: this topic is grounded in the reviewed GGUF.MiRust.com source snapshot. It documents observed code and artifacts without claiming broad deployment, model quality, or production readiness.
Build outputs
The release WASM target is wasm32-unknown-unknown. The Visual Studio build helper runs Cargo, builds the workspace, builds the runtime in release mode, and copies the generated module to tinyrustlm/runtime/tinyrustlm.wasm, the path consumed by the browser UI.
Visual Studio wrapper
TinyRustLM.sln contains an NMake project. Visual Studio does not compile Rust directly; it invokes checked-in PowerShell scripts for build, run, and test. F5 searches for a loopback port, verifies an existing server by probing the app and WASM route, or starts the local Rust server.
Reproducible release record
A release must capture Rust toolchain version, target installation, Cargo lockfile, exact commands, generated WASM SHA-256, generated model hashes, source archive hash, and whether checked-in artifacts match a clean build.
Current audit gap
The build could not be rerun in the MiRust packaging environment because a Rust toolchain was not installed. The existing WASM artifact was exercised directly through Node WebAssembly instead.
Scope
This starter page defines the questions, boundaries, evidence, and failure modes that should be recorded before a capability is presented as supported.
Engineering considerations
- Identify the source, version, target environment, and owner.
- Separate observed values from estimates and externally reported values.
- Record trade-offs, unsupported cases, and fallback behavior.
- Link performance statements to a compatible benchmark methodology.
Verification questions
- What exact artifact, revision, backend, and environment were reviewed?
- Which assumptions could change the result?
- Which data should be retained so another engineer can reproduce the conclusion?