Release build profile and artifact placement

Documents workspace version inheritance, strict lints, no-dependency packages, size-oriented release settings, WASM copy step, and checks required for reproducible release bytes.

Experimental
Last verified
2026-06-25 00:00 UTC
Updated
Reading time
2 minutes

Documents workspace version inheritance, strict lints, no-dependency packages, size-oriented release settings, WASM copy step, and checks required for reproducible release bytes.

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.

Workspace

Four crates inherit version 0.1.0, Rust 2021, proprietary license metadata, and repository identity. missing_docs is denied and unsafe code is warned at workspace level.

Release profile

codegen-units=1, full LTO, opt-level="z", and panic="abort" prioritize a compact WASM artifact over compile speed.

Build path

The runtime is built for wasm32-unknown-unknown, then copied from Cargo target output to tinyrustlm/runtime/tinyrustlm.wasm.

Reproducibility gap

The source snapshot does not pin a Rust toolchain file or record compiler/LLVM versions in the model UI. Reproducible release evidence must add toolchain identity, clean-build commands, environment, final WASM hash, and byte-comparison results.

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?