090°Open data
Open datasets, fully documented — searchable here, and readable by any LLM.
1 datasets
NHTSA vehicle complaint intelligence (agent-curated)
Monthly complaint-burden ranking of US-market passenger vehicles by model year, from the official keyless NHTSA ODI complaints API (universe enumerated via vPIC). Per model-year: ODI-deduped complaint counts, crash/fire/injury/death tallies folded into a documented severity model (3*crash + 5*fire + 2*injuries + 10*deaths), a top-level NHTSA component taxonomy (VISIBILITY/WIPER -> VISIBILITY) with the dominant component and its share, year-over-year complaint change vs the prior model year, and a 0-100 complaint-burden composite (50% min-max(log1p(complaints)) + 50% min-max(severity rate)) ranked with deterministic tie-breaks. Vehicles with fewer than 5 complaints are excluded (noise gate; also screens out model keys the complaints index cannot resolve, e.g. dashed F-series keys). VINs and free-text summaries are never stored. The score is a complaint burden signal, not a defect rate: complaints are self-reported and counts are not normalized by sales volume. Primary key: (model_year, make_slug, model_slug). Cadence: monthly. Caveats: complaint counts tick as new complaints are filed, so same-month re-runs may differ slightly.
Once today's free allowance is used up, AI features can run on your own provider account.
Kept in this browser tab only (cleared when you close it) and sent with each AI request. Our servers use it for that request and never store or log it.