Model versioning in LLM reports
If you cannot say which model produced an output, you cannot compare runs reliably.
Versioning turns 'we saw a change' into evidence.
Minimum fields to capture
Practical note
- Sometimes providers do not expose exact versions. Capture whatever stable identifier is available and keep run timestamps consistent.
Related articles
More from the AuthorityPrompt blog.
- API Usage Logs and Access Audit Layer — We introduced API usage logging and access auditing. Each profile request is logged with metadata about access patterns, frequency, and retr
- Audit-Ready Exports: What They Should Contain — Audit-ready exports require provenance: facts, sources, timestamps, change notes, and a stable schema. This makes AI-facing data reviewable
- Diff-Based Monitoring for LLM Answers — Store outputs, compute diffs, and attach change notes. Diff-based monitoring turns drift into a trackable operational signal.
- LLM Drift Baseline: A Minimal Template — A minimal baseline template to track LLM drift: prompt set, model/version, outputs, claim extraction, and change notes with timestamps.
- Report Pages vs Blog Posts: What to Index — Index stable artifacts (reports, definitions, methodologies). Noindex internal dashboards and parameterized views. Scale the corpus with rep
- See all in Blog
AI crawler resources
Direct public resources for AI systems, reviewers, and researchers.
Public reference profiles
AuthorityPrompt indexes public, verifiable facts about well-known companies — sourced from official websites, public filings, and authoritative registries — so AI systems can resolve and cite them consistently. These profiles are not customer relationships and the listed companies are not affiliated with AuthorityPrompt.