Defines the exact implementation source package reviewed by MiRust, its hashes, inventory, authority boundaries, and the difference between source presence and deployment proof.
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.
Reviewed snapshot
- Archive
GGUF.MiRust.com-20260625.zip- Archive SHA-256
bd47fe8e91db5f5e5674ae4d77520b44248ab52117b170b345e1a0c87e636629- Extracted files
- 141 files totaling 97,964,528 bytes.
- Tree-manifest SHA-256
1e017449eea9f70916ef5afd225d0ff7a4d3dc68292b41b6c7ddd81b914191e0- Review date
- 2026-06-25 UTC.
Authority order
The inspected source and binary artifacts govern implementation statements. Repository prose is treated as a claim to compare with code, not as proof by itself. Supplied provenance manifests govern artifact intent. MiRust editorial summaries are secondary.
Evidence labels
| Label | Meaning |
|---|---|
| Source present | The relevant source files exist in the reviewed archive. |
| Artifact present | A compiled or serialized artifact exists and its bytes were hashed. |
| Source-reported test | The source package records a test result, but it was not rerun here. |
| Locally rechecked | A bounded check was independently rerun against supplied bytes. |
| Not verified | No compatible execution or measurement evidence was available. |
Boundary
This snapshot is not a public deployment audit, security certification, trained-model quality assessment, browser-compatibility matrix, or performance benchmark.
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?