Open engine
The runtime is open source. Clone it, run it locally, fork it, contribute back.
The engine and schemas are open. Payment covers maintained pack work and the hosted lanes DaedalMap runs for you. Use the hosted app, download or self-host what you can, or pay for ready-to-use processing when DaedalMap is doing the work.
Everything in this list is free.
The runtime is open source. Clone it, run it locally, fork it, contribute back.
The pack schema and loc_id model are public and documented. Bring your own data in the same shape.
Every published pack is visible in the hosted catalog. Use free lanes through hosted surfaces, inspect metadata publicly, or move into the GitHub/manual path when you want to run the engine yourself.
Open the hosted app. Ask a question on the map. No account, no fee.
Small loc_id, point lookup, geometry availability, hierarchy, and boundary checks stay free with public caps. Heavy bulk processing can move to a paid lane.
Hosted lanes are paid where DaedalMap carries a real operating cost.
For humans and teams preparing data through a funded account.
Card-funded account balance
Use account credits for bulk geometry, reference conversion, BYO-data cleanup, and hosted processing jobs. The same capability can stay free for a small sample and use credits when the batch gets serious.
Sonnet up to $0.10 per question · Opus up to $1.00 per question
Research mode runs cross-domain queries through hosted LLM compute. You pay what we pay. Costs scale with query depth and corpus size.
Pick the feeds you operate on. Pay for what you watch.
$20 per month
Ops mode runs live feeds and focused operational features on continuous hosting. Monthly fee covers the running infrastructure behind active collectors and watch-state features.
Paid packs start at $0.01 per call
Programmatic callers can use wallet-native x402 payment for paid dataset calls. Account-credit access will use the same capability pricing for callers who prefer card-funded balances.
Why some things are paid
DaedalMap is not charging for ownership of public facts. It is charging for maintained pack work and hosted operation. Maintaining packs, running hosted LLM compute, operating live collectors, and keeping agent access clean all cost money. The engine stays open for anyone who wants to use the GitHub/manual path.
Programmatic access works the same way conceptually: DaedalMap prices the capability, then the caller chooses a payment rail when one is available. Agents can pay wallet-native x402 on Base USDC. Human users and teams can use account credits for hosted processing and future API/MCP usage. The ledger is shared; the payment preference changes, not the underlying tool.
Current paid dataset calls start at $0.01 for up to 100 rows, then $0.0001 per additional row, with the exact amount shown before any charge. Bulk geometry, conversion, and BYO-data processing should follow the same principle: small samples are free, serious hosted processing is metered.
The packaged local runtime sits beside these hosted lanes as a controlled research path. It is for researchers, academics, and serious teams who need more local control, larger workflows, or their own LLM keys.