> ## Documentation Index
> Fetch the complete documentation index at: https://infercrane.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Product qualification

> Run InferCrane's resumable, commit-bound local and real-infrastructure evidence system.

# Product qualification

InferCrane has one top-level qualification command. It composes the maintained unit, race,
PostgreSQL, Docker, Kind, black-box journey, provider-contract, SDK, Terraform, security,
documentation, supply-chain, and real-provider suites. It does not replace those suites or hide
their logs. Overlapping release work runs once: the black-box gate runs user journeys, while the
supply-chain gate owns archive, checksum, SBOM, and Homebrew verification.

```bash theme={"theme":{"light":"github-light-default","dark":"vesper"}}
make qualify-product
```

The command is resumable. A passed gate is reused only for the exact Git commit. The worktree must
be clean, so uncommitted changes can never inherit release evidence from an earlier source tree.
Evidence is written to:

```text theme={"theme":{"light":"github-light-default","dark":"vesper"}}
.infercrane/product-qualification/<commit>/
├── gates/
│   ├── developer-environment.json
│   ├── product-journeys.json
│   ├── adapter-contracts.json
│   └── supply-chain.json
├── report.json
└── report.md
```

Inspect the current matrix without running tests:

```bash theme={"theme":{"light":"github-light-default","dark":"vesper"}}
make qualify-product-status
```

Longer fuzz, soak, and fleet simulation run separately so pull-request feedback remains fast:

```bash theme={"theme":{"light":"github-light-default","dark":"vesper"}}
make qualify-product-nightly
```

See [Reliability test environment](/testing/test-environment) for the evidence ladder, provider
simulation boundaries, version matrix, and scheduled workflow.

The verdict vocabulary is intentionally strict:

| Status                | Meaning                                                                       |
| --------------------- | ----------------------------------------------------------------------------- |
| `PASSED`              | This exact commit passed the named gate.                                      |
| `FAILED`              | The gate ran and failed. Cleanup cannot change this result.                   |
| `NOT_RUN`             | A locally automatable gate has no evidence for this commit.                   |
| `NOT_RUN` (scheduled) | A nightly fuzz, soak, scale, or version gate has no evidence for this commit. |
| `REAL_INFRA_REQUIRED` | Fixtures prove the adapter logic, not the external provider semantics.        |
| `BLOCKED_ACCESS`      | The real suite was requested but credentials or configuration were absent.    |
| `HUMAN_REQUIRED`      | Visual, hosted, or operator-comprehension review remains.                     |

## Real RunPod qualification

Paid provider mutation always requires an explicit approval flag. Reuse the run ID after a shell
disconnect; durable operations and stage evidence allow the workflow to resume.

```bash theme={"theme":{"light":"github-light-default","dark":"vesper"}}
export RUNPOD_KEY_FILE="$HOME/.config/infercrane/runpod-key"
export INFERCRANE_RUNPOD_SERVERLESS_TEMPLATE_ID="<immutable-template-id>"
export INFERCRANE_V2_QUALIFICATION_RUN_ID="$(date -u +%Y%m%dT%H%M%SZ)-$(git rev-parse --short HEAD)"

./scripts/qualify-product.sh runpod --approve-paid-resources
```

The command records four independent gates: serverless lifecycle, serverless faults, elastic
lifecycle, and elastic faults. A provider allocation failure in one gate cannot erase successful
evidence from another. Together they cover vLLM protocols, AIPerf, autoscaling, Release Guard,
streaming cancellation, lost-response adoption, disruption, deletion, and direct zero-inventory
checks. A provider allocation or image-transfer failure remains a failed suite even when guarded
cleanup succeeds.

## AWS, GCP, and Kubernetes GPU qualification

The portable qualification runner needs an isolated provider environment, immutable workload specs,
and one API key shared by the control plane and workers. See [provider setup](/provider-setup) before
running any paid gate.

```bash theme={"theme":{"light":"github-light-default","dark":"vesper"}}
export INFERCRANE_V1_PROVIDER_ENV_FILE=/private/provider.env
export INFERCRANE_V1_SPEC_DIR=/private/qualification-specs
export INFERCRANE_V1_API_KEY_FILE=/private/control-and-worker-key

./scripts/qualify-product.sh aws --approve-paid-resources
./scripts/qualify-product.sh gcp --approve-paid-resources
./scripts/qualify-product.sh kubernetes --approve-paid-resources
```

The compatibility-prefixed `INFERCRANE_V1_*` names do not select an old release. A gate remains
`REAL_INFRA_REQUIRED` until its exact-commit paid run passes; the existence of the guarded runner is
not qualification evidence.

## What the system proves

Local qualification proves control-plane behavior against deterministic dependencies, a real
PostgreSQL process, Docker process failures, and a real Kind API. It cannot prove GPU availability,
provider eventual consistency, customer networking, runtime performance, public package delivery,
or human usability. Those boundaries remain separate rows in `report.json` so automation can enforce
release policy without turning simulated evidence into a production claim.