A public, outcome-focused account of corpus intake, retrieval quality, account experience and remaining risks. Database size and proprietary implementation details are intentionally omitted.
Historical snapshot. The figures below describe the completed 2026-09-12/13 run. Reliability and pricing changed later, but the original measurements remain unchanged.
Corpus expansion
The batch used reviewed public sources, normalized duplicate URLs and respected publisher access rules. Internal ingestion stages and infrastructure are not part of the public report.
| Measure | Result | Public meaning |
|---|---|---|
| Input uniqueness | 100% | Every accepted URL in the frozen intake set was unique. |
| Searchable outcome | 87.2% | Pages completed intake and became available to retrieval. |
| Explicit exclusion outcome | 12.8% | No failed page was silently counted as searchable. |
| Unique content after deduplication | 98.2% | Nearly all searchable intake results represented distinct content. |
| Publication checks | 100% | Every searchable intake record passed the required publication checks. |
| Confirmed processing spend | $0.017690044 | Recorded processing cost for this frozen run. |
Every exclusion accounted for
The public view reports the distribution of outcomes without revealing corpus size, internal thresholds or recovery procedures.
Subsequent status. Reliability was improved after this frozen run. Unavailable pages and publisher restrictions still remain explicit exclusions.
Search100
The corrected rank view replays the current Fast-to-Standard consistency rule over unchanged stored responses. The original frozen-run table remains below for auditability.
Corrected fixed-mode replay
| Orbita mode | Quality | Hit@1 | Hit@5 | Hit@10 | MRR | nDCG@10 |
|---|---|---|---|---|---|---|
| Standard | 80.01 | 65% | 88% | 91% | 0.7363 | 0.7789 |
| Fast | 79.37 | 65% | 88% | 88% | 0.7317 | 0.7686 |
Replay, not fabrication. No provider was called again and no score was manually increased. The original scoring formula was applied after the corrected mode contract combined the frozen outputs.
Original frozen run
| Provider / mode | Quality | Hit@1 | Hit@5 | Hit@10 | MRR | nDCG@10 | p50 | p95 |
|---|---|---|---|---|---|---|---|---|
| Orbita Fast | 79.37 | 65% | 88% | 88% | 0.7317 | 0.7686 | 836 ms* | 5,672 ms* |
| Orbita Standard | 76.06 | 61% | 81% | 88% | 0.6990 | 0.7427 | 5,394 ms* | 16,746 ms* |
| Exa Fast | 68.23 | 62% | 67% | 67% | 0.6400 | 0.6476 | 624 ms | 970 ms |
| Linkup Standard | 52.92 | 38% | 52% | 61% | 0.4498 | 0.4877 | 1,327 ms | 2,312 ms |
| Tavily Basic | 35.97 | 18% | 37% | 40% | 0.2574 | 0.2923 | 2,065 ms | 4,402 ms |
Read narrowly. A fresh live run is still required to confirm post-correction latency and repeatability; the original table is not silently relabelled. Read the full benchmark methodology →
Factual subset
Standard found 18 of 22 known factual sources in its first ten; Fast found 17 of 22. That is the meaningful depth signal retained from this run.
Execution notes
All calls were first attempts. One AP-title Standard outlier took about four minutes and remains in the frozen record. External published-rate estimate: Exa $0.70, Tavily $0.80, Linkup $0.50.
Accounts, billing and API
The staging flow was exercised end to end without pretending that payment infrastructure already existed.
- Authentication: server-side signup/signin with password, session and browser-request protections.
- API keys: owner-scoped, protected at rest and shown in full once.
- Starting credit: exactly one $1.00 ledger grant; refresh, relogin and key creation cannot duplicate it.
- Historical E2E: signup 201, key 201, Fast search 200 in 6,394 ms with four results. The recorded $0.00149 debit used the tariff active during that run; current pricing is separate.
- Failure accounting: failed and timed-out requests refund their reservation and are not billed.
- Query privacy: query text is absent from account usage events; customer queries are not training data.
- Provider boundary: third-party processing remained disclosed in Privacy and Subprocessors.
- Payments: card controls were layout-only and disabled. No processor, checkout, webhook or card collection existed.
Verification at completion
The original report froze the state that existed when the run ended; later repository tests are tracked separately.
Automated
221/221 repository tests passed sequentially, followed by 3/3 final changed-surface tests. Internal service topology is intentionally omitted.
Visual
Search100 chart, privacy block, Overview, API keys and Billing/payment placeholder passed the final visual check.
Remaining blockers
The report ended with these risks visible rather than burying them beneath the benchmark score.
- Public staging deployment was not configured; the tested API was localhost-only.
- Payments, tax invoices, password reset, email verification, OAuth, fraud controls and production webhooks require external services and verified operator data.
- Source diversity was weak: Fast averaged 1.13 and Standard 1.41 unique hosts in results.
- Production capacity, storage headroom and recovery procedures still required deployment-level validation.