Comparison coverage
7 comparisons across 5 provider clusters.
- Two-sided verified: 5
- One-sided verified: 2
- Pending: 0
Workspace
AI Model Infrastructure Intelligence Workspace
Operator view across the entity graph — verified models, provider coverage, pricing references, source freshness, comparison clusters, and the manual reverification queue in one snapshot. The same verified-data backbone that powers the AI usage learning curriculum at /learn. Every count below is derived from the typed local data layer. No fabricated metrics, no rankings, no auto-mutation.
Navigate
Workspace destinations
Every hub the catalogue exposes, with a one-line description of what lives there.
Decision briefs
Export source-backed evidence packs from selected models. Markdown or JSON.
Guided demos
Three predefined route plans that walk the full workflow on real verified data.
Model catalogue
Every tracked model with verification status, pricing references, and lifecycle.
Provider coverage
Model creators and hosting platforms with primary docs URLs and verification.
Pricing references
First-party and hosted pricing references with freshness chips. Not live quotes.
Comparison clusters
Side-by-side reference pages — never a price-ranking or winner engine.
Source registry
Every primary-source citation with retrievedAt and freshness state.
Coverage map
Verification breadth across every entity domain, plus the reverification summary.
Status observations
Vendor-reported status feed plus independent host probes. No uptime claim.
Reverification queue
Manual review queue for sources that have aged out. No auto-mutation.
Research guides
Methodology articles behind each verified field.
Technical docs
Reference schemas, integrity rules, and field definitions.
Snapshot
Current intelligence snapshot
Counts derived from the typed local data layer. Cards link to the corresponding hub view.
Operations
Review operations
The high-priority subset of the reverification queue, grouped by reason. The queue is informational — never auto-mutated.
Health
Coverage health matrix
Per-domain readiness across the entity graph. There is no derived health score — counts are what they are.
| Domain | Verified / available | Partial / review due | Blocked / missing | Primary action |
|---|---|---|---|---|
| Models | 9 | 2 | 1 | Walk model cards in /models; review null metrics |
| Providers | 7 | 1 | 0 | Confirm primary docs URLs in /providers |
| Pricing | 44 | 0 | 0 | Re-verify first-party pricing; references are not live quotes |
| Hosted availability | 2 | 0 | 0 | Confirm hosted model IDs and rates |
| Comparisons | 5 | 2 | 0 | Two-sided verified comparisons are indexable |
| Status | 2 | 2 | 0 | Status pipeline reports vendor-reported observations only |
| Sources | 14 | 0 | 0 | Re-fetch citations on the reverification queue |
| Reverification | 0 | 11 | 6 | Work the queue manually; never auto-update |
Citation impact
16 unique primary-source citations support the verified entity graph.
- Fresh: 16
- Review due: 0
- Stale: 0
Methodology
How these counts are produced
Model selection
What to weigh when picking a model.
Source verification methodology
Primary-source rules, freshness lifecycle, and the manual review loop.
API pricing methodology
References, not live quotes. No price ranking.
AI provider status monitoring
Vendor-reported feeds + independent probes. No fabricated uptime.
Data verification rules
Verification states, freshness lifecycle, stale ≠ false.
Comparison methodology
How comparisons render side-by-side.