AI Governance Institute
All governance templates →How do we intake and govern open-weight and self-hosted AI models?

Implementation Kit

Open-Weight AI Model Intake and Deployment Checklists

The controls that only apply when you download and self-host: license review, weight integrity verification, deployment configuration capture, and the pre-deployment testing no vendor did for you. An intake checklist, a config template, an integrity log, and an update policy.

Who this is for: The platform or ML owner bringing an open-weight or self-hosted model into production.

Download the kit (Markdown) ↓4 artifacts. Every table also copies as CSV.

1. Open-weight model intake checklist

Spreadsheet

The gate for a self-hosted model. License, security, integrity, and classification before anything runs.

Template

CheckStatusEvidence
License reviewed; permitted uses and redistribution terms understoodY / N
Model card / release notes reviewed for known limitations and evalsY / N
Weights downloaded from the official source over a verified channelY / N
Integrity verified against a published checksum or signatureY / N
Security assessment: known CVEs, unsafe deserialization, bundled codeY / N
Pre-deployment adversarial testing run (prompt injection, jailbreak, harmful content)Y / N
Risk tier assigned via the classification methodY / N
Named owner for security-disclosure monitoring assignedY / N

Worked example

CheckStatusEvidence
License reviewedYApache-2.0; commercial use OK; attribution noted
Model card reviewedYWeak non-English coverage; eval suite dated
Weights from official sourceYDownloaded from the vendor HF org
Integrity verifiedYSHA-256 matched the published manifest
Security assessmentYsafetensors format; no pickle; no CVEs open
Adversarial testing runPartialprompt-injection + jailbreak done; harmful-content suite pending
Risk tier assignedYLimited (internal drafting, no personal decisions)
Disclosure-monitoring ownerYML Platform on-call
GateBlockedcomplete the harmful-content suite

Acceptance criteria

  • No self-hosted model serves traffic before every check passes or a named approver logs a time-boxed exception.
  • Adversarial testing is run in-house, since no vendor has done it.
  • The license review names the specific license and its commercial and redistribution terms.

2. Deployment configuration template

Spreadsheet

A record of exactly what is running: which weights, at what precision, behind what guardrails, with what fine-tuning.

Template

FieldEntry
Model and version
Source and download URL
Published checksum / signature
Computed checksum at deploy
Quantization / precision
Serving stack and version
Guardrail configuration (input filter, output filter, system prompt)
Fine-tuning applied (dataset, method, date)
Inference infrastructure and isolation
Rollback target

Worked example

FieldEntry
Model and versionOpenLM-2 13B, release 2026-07
Sourcevendor Hugging Face org, tag v1.0
Published checksumsha256:4b91... (manifest signed)
Computed at deploysha256:4b91... (match)
Quantization8-bit (bitsandbytes)
Serving stackvLLM 0.6.x
Guardrailsinput PII filter; output harmful-content filter; system prompt v4
Fine-tuningLoRA on 12k internal support transcripts, 2026-08-05
Infradedicated GPU node pool, no egress to public internet
Rollback targetprevious LoRA adapter v3

Acceptance criteria

  • The computed checksum at deploy is recorded and matches the published one.
  • Every guardrail and any fine-tuning is captured, with dates.
  • The config is version-controlled and updated on any change to weights, quantization, or guardrails.

3. Weight integrity verification log

Spreadsheet

A running record that every set of weights placed in the artifact repository was verified.

Template

ModelVersionDownload datePublished checksumComputed checksumMatchVerified by
<model><version>YYYY-MM-DD<hash><hash>Y / N<name>

Worked example

ModelVersionDownload datePublished checksumComputed checksumMatchVerified by
OpenLM-2 13Bv1.02026-07-30sha256:4b91...sha256:4b91...YML Platform
OpenLM-2 13Bv1.12026-09-02sha256:7c02...sha256:7c02...YML Platform

Acceptance criteria

  • Every set of weights in the repository has a log entry with a match result.
  • A mismatch blocks use and triggers a re-download from the official source.
  • The artifact repository has access controls and audit logging separate from this log.

4. Update and retirement policy

Document

How often you evaluate newer releases, what triggers an upgrade, and how a self-hosted model is retired.

Template

One page.

Evaluation cadence: review newer releases of the deployed model family every <N> months, and within <N> days of a release flagged for a security fix.

Upgrade triggers: a security advisory affecting the running version; a new release with a material capability or safety improvement on our eval suite; a license change.

Upgrade process: run the full intake checklist for the new version; compare on the internal eval suite; stage before production; keep the prior adapter as rollback.

Retirement: delete weights from serving infrastructure and the artifact repository per the data policy; archive the deployment config, eval history, and integrity log; update the registry and inventory; owner sign-off.

Worked example

OpenLM-2 update log:

  • 2026-09-02: v1.1 released with a tokenizer security fix. Intake checklist re-run; eval delta negligible; promoted to production 2026-09-04; v1.0 adapter kept as rollback for 30 days, then archived.
  • Next scheduled family review: 2026-12.

Acceptance criteria

  • The evaluation cadence is a specific interval, and a security release short-circuits it.
  • An upgrade re-runs the full intake checklist, not a subset.
  • Retirement removes weights from both serving infra and the artifact repository.

Governance controls this kit produces evidence for

Completing the artifacts above gives you a head start on the evidence requirements for these controls.

SCT-006
Self-Hosted Open-Weight AI Model Governance

The whole kit is the self-hosted open-weight model governance process: intake, config, integrity, updates.

DGC-006
AI-Generated Code and Open-Source License Compliance

The license-review step of intake covers open-source license compliance for model artifacts.

SEC-002
AI System Access Controls

The access-controlled artifact repository and infra isolation fields evidence AI system access controls.

SEC-005
Adversarial Robustness Testing

The pre-deployment adversarial testing step is the adversarial robustness testing record.

MGV-002
AI System Intake and Approval Workflow

The intake checklist is the approval gate for self-hosted models entering production.

This kit backs one playbook. Read the full guidance for the reasoning behind each artifact.

Decide what to implement next

Assess your governance gaps, then create an action plan with owners and target dates. Build and export without an account; sign in when you want to save your plan.

Start the AI governance assessment →