How can you compare OpenAI and Gemini fanout?

Run the same topic once with each provider, keep locale fixed, save both results and compare them locally. The browser shows exact query overlaps and each run side by side without another API request.

Use this before comparing providers or the same provider on different dates.

Compare OpenAI and Gemini locallyRun the same topic once with each provider, save both observations and compare their exposed query strings on your device.Start the model comparison

Set the comparison question and scope before selecting runs

A provider comparison examines two providers under a documented setup. A repeat examines two observations from one provider. A mixed comparison changes several conditions and cannot separate their effects. Local history labels comparison type from topic and provider; it does not automatically verify every locale, model and protocol field.

Review input, language, country, model label, timestamp, tool version, method and status yourself. Source scope and retention filters belong in the same matrix. Name differences before interpreting query overlap. The intersection tells you which normalized strings match, not which provider produces better answers.

For the next step, use the explicit test protocol.

Our completed matrix uses two previously approved OpenAI observations. It is repeat analysis, not an executed OpenAI-versus-Gemini benchmark. Reuse the columns for your own permitted provider comparison, without replacing absent provider fields with assumed values.

Read the two OpenAI observations in the matrix

The records are best-seo-tools-us and best-seo-tools-repeat. Both use best SEO tools, en/US and the gpt-5.6-luna model label. The matrix retains their exact 27 August 2026 UTC timestamps, native-fanout-tool/1.0.0 and provider-native-search/1.0. Both responses are incomplete; these data do not justify a complete answer comparison.

See the context in the dated repetition case.

Each observation records four query strings and one search action. Exactly one string matches after NFKC, trimming and lowercasing. The domain sets contain eleven and ten entries. Those counts describe visible data volumes. More sources establish neither better research nor higher quality, and a shared query string does not establish an equivalent answer.

The matrix's final group names missing evidence: full answer text, a reconstructable sentence-to-citation mapping and a second provider run in the public corpus. Published domains retain their documented scope rather than being assigned to individual queries afterwards. These gaps are part of the result and travel with the report.

Compare zero, failure and removed sources correctly

A completed run can expose zero query strings. A failed request is not a successful zero observation. An incomplete run with visible strings remains incomplete. Review status in the full native export: query selection omits providerResponseStatus and cannot establish that status by itself.

For the next step, use the selection and full-export exercise.

Gemini grounded sources and Search Suggestions are removed before local saving. An empty saved source list therefore does not have the same evidence basis as a permitted current OpenAI source list. Compare permitted query strings where useful, retain the filter notice and leave the incompatible source assessment open.

Local history holds up to 20 entries for no more than 30 days in this browser and origin. Export permitted full comparison records with metadata if you need the working record longer. The schema guide separates export families; the reporting template retains the question actually answered and its proof boundaries.

For the next step, use the reporting template.

Common questions

These short answers cover recurring questions from tool use and source review. They are not claimed search-volume data.

Is an OpenAI-versus-Gemini result a benchmark?: No. Two saved runs show what those API calls exposed under one setup. They do not establish general model quality, consumer-product behavior or a statistically reliable winner.
When is it a provider comparison and when a date comparison?: The tool labels equal topics across different providers as a provider comparison. Equal topics from the same provider on different dates form a date comparison; mixed selections are labelled accordingly.
Why is Gemini evidence not part of the public corpus?: The current rights boundary permits browser-local use but does not establish publication rights for retained Gemini Grounded Results or Search Suggestions. Public examples therefore remain limited to reviewed OpenAI owner-run observations.

Sources used

These sources support the functions and limits described here. They do not prove claims beyond that scope.

Original annotated workflow, controls, prompts and export exercises
Original diagrams and editorial annotations; only the previously approved normalized OpenAI records. Synthetic selection files use actual producers and independent CSV readback, not new provider output or Excel/Sheets import. Own dated public-page checklist makes indexing and outcomes explicitly unproven.
Open source
Current website export contracts and reference validation
Draft 2020-12 reference for the current native/modelled full runs, query selections and local comparisons. Schema checked against actual service/export functions with synthetic provider stubs and semantic negative tests. No root schemaVersion is invented for full runs. Legacy planner and handoff formats remain separate.
Open source
OpenAI web search tool guide
Documents web_search_call output and search actions that usually, but not always, include the query or queries searched. Search actions, exposed strings and cited sources therefore need separate interpretation. It does not expose chain of thought or guarantee that every route returns query strings. Rechecked for the original 2026-10-10 zero-query decision table; no new provider observation is published.
Open source
Gemini grounding with Google Search
Documents Gemini 3.7 Flash Google Search, google_search_call arguments.queries, cited URL annotations and per-query billing.
Open source
Gemini API Additional Terms
Restricts caching, syndication, analysis and reuse of Google Grounded Results and Search Suggestions; drives the Gemini local-save redaction and public-example boundary.
Open source
Direct OpenAI and Gemini native-fanout implementation
One direct provider request, strict provider allowlist, provider-exposed query/source parsing, honest zero-query output, CAPTCHA-first shared atomic quota and budget accounting, no raw server archive and separate public flag. Bounded live production fixtures passed for OpenAI and Gemini on 2026-08-26.
Owner record
Opt-in browser-local result history
Explicit save action, IndexedDB-only storage, maximum 20 entries, 30-day expiry, individual deletion, clear-all and local JSON/CSV export without accounts or shared response caching.
Owner record
Dated OpenAI fanout example observations
Four owner-run OpenAI API observations record exact inputs, timestamps, exposed query strings, search-action source domains, usage, method versions and response status. Versioned JSON, normalized CSV and a JSON Schema are published. They are not an independent benchmark.
Open source