Not one of the seven search APIs we tested publishes a deprecation policy with a notice period. No vendor scores above 2 of 4 on documentation and commitment. Separately, we measured retrieval against 1,000 SimpleQA questions: the fastest provider ranked last, and the best retriever was fifth of seven on speed.
What each vendor publicly commits to, scored 0–4 on five fixed axes. Six vendors.
Scored disclosures, not measurements. A low score means poor disclosure, not poor practice. No confidence intervals apply. Compliance is the weakest axis — several scores rest on a trust center existing rather than a report being retrievable, so read the source grade on each cell.
Verified 2026-09-03 · valid until 2026-12-02
1Price stabilityHow much a buyer's cost basis has moved, and how existing customers were treated when it did.
2Supply independenceWhether the vendor can be cut off or repriced by an upstream it does not control.
3Documentation and commitmentWhether the vendor has made the operational commitments a production integration depends on.
4ComplianceWhether the vendor clears the first filter of a standard vendor security review.
5Procurement maturityWhether an organisation with a procurement function can actually buy this, and how much friction it will meet.
Exa
founding date unresolved
Price stabilityconflict2/4
Base search moved $5 to $7 per 1,000 inside the window. Full pricing public. The transition date is not established: the 2026-03-03 changelog announces simplified and lowered pricing and fixes the $7 level on that date, it does not date the rise. Offsetting reductions on 2026-03-03 and 2026-07-14. Material effective-price-drift exposure.
Changelog dated 2026-03-03 announces simplified and lowered pricing: contents bundled free for the first 10 search results, search with contents $7/1,000, additional results $1/1,000, summaries $1/1,000, Deep $12/1,000, new Deep Reasoning tier $15/1,000, applied automatically with no action needed. Establishes the price level on that date, not the date the level changed.
The rise is real but undated. Re-derive from Internet Archive captures of exa.ai/pricing between the 2025 $5 citation and 2026-03-03. See conflict exa_increase_date, which now carries three competing dates.
Flagged on the cell regardless of score: surcharge above 10 results, per-content-type billing, metered Agent runs against a $5 default cap, and Connect provider passthrough.
Base Search raised to $7 per 1,000 here — competing record, 2026-07-29
Unresolved. None of the three positions establishes when the rise happened: the changelog fixes a level and the two tracker dates postdate it. The transition must be re-derived from Internet Archive captures of exa.ai/pricing between the 2025 $5 citation and 2026-03-03. Does not change the score, which rests on the rise having occurred inside the window.
Source conflict — Exa Agent Compute Unit pricing
ACU moved from $0.0001 to $0.10 in the same change that lowered per-run caps — usagepricing, 2026-07-14
Concurrent reductions in the same 2026-07-14 change: free tier 1,000 to 20,000 requests/month, Agent range $0.025-$2.00 to $0.012-$1.00 with a new Minimal mode, X-high halved to $1.00, in-agent search $0.007 to $0.005 — usagepricing, 2026-07-14
A 1000x unit change alongside concurrent reductions across every other line reads as a units redefinition rather than a price move. Kept out of the increase count. Status moves from unresolved to substantiated once the concurrent reductions are confirmed from a primary source.
Supply independence4/4
Operates its own neural index, corroborated by third-party analysis of result composition and by presence of a named crawler in public AI-crawler registries.
No recorded rate change in the window. Earliest preserved pricing capture matches the current structure. Pricing stable through the Nebius acquisition.
no_adverse_change_24mObservedThird-party reported
Wayback-derived. Verify against live pricing page before publication.
No adverse rate change in the window. Recurring monthly credit. Downgraded from 4 by a long-standing documented discrepancy between the marketing page and billing docs on deep search, closed 2026-07-14, and by a new Extract endpoint with undisclosed variable pricing.
SOC 2 Type II and ISO 27001 claimed, trust center exists, BYOC available. Certificates not independently verified and DPA availability not confirmed on accessible public pages.
soc2_type_iiObservedVendor-claimed
Bullet-point claim. Auditor and report not identified.
No adverse change in the window. Two documented decreases and two free-tier expansions, existing customers honoured, full pricing public, which is anchor 4. Held at 4 unless the reported gap between the pricing page and the Search API docs over the 100-calls-per-day quota is confirmed against the live docs, in which case anchor 3 applies: a documented pricing-page-versus-billing-docs discrepancy triggers it regardless of which way it favours the buyer.
2025-10. API platform launch: Web Search $6.25/1,000 for 1-50 results and $8.00/1,000 for 50-100, Contents $10.00/1,000 pages. Tracker and Archive sourced; must be verified before publication.
2026-03-03. Research API launches; its rates have not moved since.
Source URL unresolved () · retrieved 2026-09-03
price_decreaseObservedDirectly verifiable
2026-03-11. Web Search consolidates to a flat $5/1,000 for up to 100 results and Contents drops to $1/1,000 pages, effective immediately, billed at the new rates from 2026-03-12. This is a reduction.
Existing pay-as-you-go and grandfathered monthly and yearly plans moved automatically to the lower rates. Enterprise and custom contracts honoured through end of term. Existing credits carried over.
By the 2026-07-29 capture, Search is free for the first 100 calls per day and $5/1,000 thereafter.
Source URL unresolved () · retrieved 2026-09-03
Source conflict — You.com Search API rate
Web Search $6.25 per 1,000 for 1-50 results and $8.00 per 1,000 for 50-100; Contents $10.00 per 1,000 pages — tracker and Archive captures, 2025-10
$5 per 1,000 Web Search, $1 per 1,000 Contents pages — you.com blog, 2026-03-11
$4 per 1,000 Smart Search, $50 per 1,000 Research, 300-query trial — costbench, 2026-04-23
Vendor primary source used. The trackers lagged an announced reduction: the 2026-03-11 change lowered Web Search to a flat $5 per 1,000, and tracker snapshots from 2026-04 to 2026-06 had not caught up.
Supply independence1/4
Aggregates third-party sources and does not disclose index provenance.
Automatic data purge claimed; scope not documented.
Source URL unresolved () · retrieved 2026-09-03
iso_27001Not disclosedNot yet retrieved
Source URL unresolved () · retrieved 2026-09-03
Procurement maturityvendor-claimed2/4
Enterprise tier and custom contracts exist. No marketplace listing, no verifiable compliance artifacts behind the enterprise path, customer claims unsubstantiated.
The search API is a monetisation surface for a browser business rather than the company's core product.
Price stabilityconflict1/4
Free tier live since May 2023 at 2,000 queries/month, raised to 5,000 under the August 2025 AI Grounding update, removed February 2026 and replaced by $5 in monthly credits worth roughly 1,000 searches, billed at $5 per 1,000 requests. The same event added an attribution requirement that did not previously apply, which is a second adverse element under section 2.3. Existing free-plan subscribers at 2,000 queries/month reportedly retained access, which is partial grandfathering; see anchor_gap.
Free tier available since May 2023 at 2,000 queries/month; raised to 5,000 queries/month in August 2025 under the AI Grounding update.
Source URL unresolved () · retrieved 2026-09-03
free_tier_removalObservedThird-party reported
February 2026. Replaced by $5 in monthly credits, roughly 1,000 searches, billed at $5 per 1,000 requests. TODO_BRAVE_ANNOUNCEMENT_URL and the full ISO date are required by the conventions and are still missing.
A restriction that did not previously apply: the Brave Search API must be attributed on the project's site or about pages. Counted as a second adverse element of the February 2026 event under section 2.3, not as a separate event.
Partial. Existing free-plan subscribers at 2,000 queries/month reportedly retained access. Regraded from observed_absent, which overstated what the sources support. Must be checked against Brave's own announcement.
See conflict brave_pricing_trackers. Two genuine tracker errors remain after endpoint scope is accounted for. Must be verified against the live page before publication.
Search $5 per 1,000 requests; Answers $4 per 1,000 plus $5 per million tokens; per-query rates $0.003 to $0.005 depending on endpoint — vendor announcement, 2026-02-12
Base plan around $3 per 1,000 — apicostcalc, 2026-08-15
$4-5 per 1,000 — costbench, 2026-08-17
Pricing entirely sales-led; no self-serve tiers published — xpay, 2026-05-03
Free forever plan still listed — apicostcalc, 2026-08-17
Vendor primary source used. Two of the three original tracker positions were endpoint scope rather than conflict and are reclassified. Two genuine tracker errors remain and are published: the sales-led claim and a free-forever listing that postdates the tier's removal.
Supply independencevendor-claimed4/4
Operates its own independent index, built without reliance on Google or Bing. Named crawler present in public AI-crawler registries.
Zero data retention offered to enterprise customers only, DPA governs standard data use, GDPR handling documented via DPO contact. No SOC 2 or ISO certification surfaced.
Thin disclosure here reflects company age, not a deliberate choice. See missing_data_policy and score_state.
Price stabilityinsufficient history
Publicly available for approximately two weeks at the verification date. No pricing history exists to assess. Scoring this as 0 would be a category error.
No artifacts recorded for this cell.
Supply independencevendor-claimed3/4
Claims its own independent index of more than 100 billion documents. No external corroboration available.
Base search rate over the trailing 24 months. Two vendors never moved.
Base search rate per 1,000 requests, 2024-09 to 2026-09 · verified 2026-09-03
A dot is a dated, sourced change
A ring marks a change our sources disagree about
An open marker means our record begins there, not the price
A dashed run means both levels are attested but the change date is not
Vendors on the same rate are drawn a hair apart — read the value from the label
6 notes on sourcing
Tavily. Billed per credit: $0.008 per credit is $8.00 per 1,000 credits. The $7.50 in the retrieval table is per 1,000 searches — a different unit, not a conflicting figure. Wayback-derived; verify against the live page.
Exa. The 2026-03-03 changelog announces simplified and lowered pricing and fixes the $7 level on that date; it does not date the rise. Two tracker records place the raise at 2026-04 and 2026-07-29, both after the changelog already quotes $7, so neither can be the transition either. Pending re-derivation from Archive captures.
Brave. The February 2026 event carries two adverse elements, not one: the tier removal and an attribution requirement that did not previously apply. Existing free-plan subscribers at 2,000 queries/month reportedly kept access, so grandfathering is partial rather than absent — that sits between two anchors and is recorded in the scorecard rather than carried by the score.
You.com. The API platform launched 2025-10, so there is no earlier level to plot. Existing pay-as-you-go and grandfathered plans moved automatically to the lower rates, enterprise contracts were honoured through end of term and credits carried over. Launch rates are tracker and Archive sourced and still need verifying.
Keenable. Publicly available roughly two weeks at the verification date. The short line is insufficient history, not a stable record.
Brave, Tavily and Linkup rest on third-party and archive-derived captures rather than a live vendor page, and the scorecard flags all three for direct verification before publication. Nothing between two marked points is measured — a flat segment means no change was recorded, not that one was ruled out.
Every dated base-rate event in the 2024-09-01 to 2026-09-03 window, 10 in total. Nothing is interpolated: a vendor appears here only where the date is established in the published evidence.
Date
Vendor
Rate per 1,000
What changed
Source grade
2024-09-01
Brave
free tier
Free tier at 2,000 queries/month, live since May 2023
Third-party reported
2024-09-01
Exa
$5.00
$5.00 per 1,000, from a 2025 forum comment by a team member
Vendor-claimed
2024-09-01
Linkup
$5.00
€5 to $5.00 per 1,000 — currency only, no rate change
Third-party reported
2024-09-01
Tavily
$8.00
$0.008 per credit, unchanged through the window
Third-party reported
2025-08-01
Brave
free tier
Free allowance raised to 5,000 queries/month under the AI Grounding update
Third-party reported
2025-10-01
You.com
$6.25
API platform launch: Web Search $6.25 per 1,000 for 1–50 results, $8.00 for 50–100
Third-party reported
2026-02-12
Brave
$5.00
Free tier removed, replaced by $5 in monthly credits; $5.00 per 1,000 metered, plus a new attribution requirement
Third-party reported
2026-03-03 (change undated, after 2025-01-01)
Exa
$7.00
$7.00 per 1,000 established by the changelog. The date of the rise is not established — it happened somewhere in this span
Directly verifiable
2026-03-11
You.com
$5.00
Consolidated to a flat $5.00 per 1,000 for up to 100 results — a reduction, effective immediately
Directly verifiable
2026-08-20
Keenable
$4.00
$4.00 per 1,000 at public launch
Third-party reported
Retrieval, latency and cost
Seven providers, the same 1,000 SimpleQA questions, k = 10. Latency measured in a separate pass at concurrency 1.
1,000 search tasks
Benchology Benchmark 01
Web Search Tools
Retrieval accuracy (URL-exact match) across the same 1,000 search tasks.
Higher accuracy is better
Accuracy
70%75%80%85%90%95%
91.1%
90.0%
89.1%
84.8%
84.4%
83.4%
75.9%
Exa
Brave
You.com
Linkup
Serper
Tavily
Keenable
Search provider
● Same tasks● Same inputs● Observed resultsBar color = provider brand
Does the returned list contain the specific page a human researcher cited?
Retrieval, latency and price by provider · prices verified 2026-09-03
⚠ Serper's rate is third-party sourced, not read from a live vendor page. Treat it as indicative.
2 notes on these figures
Serper's price. Its public pricing page was not reachable and its credit packs sit behind signup. It is also the figure that most changes the comparison, since it would make Serper the cheapest per call by a wide margin.
Pending figures.pending marks a value named in the published plan but not yet in the repository. Per-provider domain rates and the latency distribution arrive with retrieval-summary.json and latency.csv.
Exact source rate is a floor: SimpleQA's reference URLs are not exhaustive, so a provider returning a different but perfectly good source scores zero. At domain level all seven fall within 2.2 points of each other.
Detail, method and caveats
Scope
Two separate studies over the same set of providers, under different evidence regimes.
Retrieval quality — measured
One query in, one ranked list of ten out, scored against SimpleQA's reference source URLs. n = 1,000 per provider, 7,000 calls, 6,998 succeeded. 95% Wilson intervals. Providers whose intervals overlap are not distinguishable at this sample size. Seven providers.
Reliability disclosure — reviewed
What each vendor has publicly committed to across five dimensions. A structured document review, not a measurement, so no confidence intervals apply and none are shown. Six providers.
What we did not measure
Answer synthesis, content extraction, multi-step agentic retrieval, index freshness, non-English or non-US performance, and whether any vendor's compliance controls actually work.
What the shapes show
No vendor is strong everywhere. The largest total area belongs to Linkup at 3-3-2-3-3, which leads no single dimension.
Compliance and independence trade off. Tavily scores highest on compliance and procurement while running no general index of its own. Brave runs one of the few genuinely independent Western indexes and scores 1 on compliance.
Documentation is uniformly weak. The highest score in the field is 2 out of 4. The artifact count that produces it is a straight tally of eight checkable items, making this the most objective axis and the most damning.
Serper was scored and removed from the scorecard. It scored at the floor on all five dimensions, and a single all-zero row compresses the visual range without adding information. Removing an outlier is defensible on presentation grounds; it also narrows the spread across the remaining six, which flatters them. Its full scored row ships in scorecard.json, and it is measured in retrieval above.
Latency
The finding that changed our own numbers.
Our first latency figures came from the main concurrent run. We re-measured with exactly one request in flight and the numbers moved by 1.5× to 7.2× — and not uniformly. The inflation tracked the per-provider concurrency we had assigned to respect each vendor's rate limits, and it reordered the ranking: the provider that looked fastest under load was second slowest in isolation, and the reverse.
In-run timings measure your own worker allocation, not vendor speed. We discarded ours. Anyone running a similar test should.
Method. Separate pass, exactly one request in flight globally, 150 questions, 1,050 sequential calls, providers visited round-robin in shuffled order. Measured from a single location on a single day — relative comparisons under identical conditions, not absolute figures reproducible from another network.
Tails. Means understate the spread. Serper's p95 is 4,911 ms, about 3.5× its median, where the other six sit near 1.5×.
Price
List price per 1,000 search requests, verified 2026-09-03. All seven bill per request.
Checked against observed spend where possible. Exa reports costDollars on every call; measured across 1,000 calls it came to exactly $7.00 per 1,000, matching its advertised rate. Linkup was checked by differencing its credit balance: $5.63 observed against $5.00 advertised, a gap caused by our own retry behaviour rather than a pricing discrepancy. Announced prices held up where they could be checked, so announced prices are what we publish.
One figure is weaker than the others. Serper's public pricing page was not reachable and its credit packs sit behind signup, so its rate comes from third-party sources. It is also the number that most changes the comparison, since it would make Serper the cheapest per call by a wide margin. Treat it as indicative.
The same 24-month window, per vendor
Price and free-tier movement · verified 2026-09-03
Vendor
Rate changes
Free-tier changes
Net on base search
Grandfathered
Exa
1 increase (undated), 2 decreases
3, all loosening
$5 → $7 (+40%)
Auto-applied
Brave
1 restructure
2, one raise then removal
free → $5/1k
Partial
Tavily
No change in 24 months — $0.008/credit throughout
n/a
Linkup
No change in 24 months — €5 → $5 /1k (FX only)
n/a
You.com
No adverse change in 24 months — 2 documented decreases, $6.25 → $5/1k
Yes, explicitly
Keenable
no history
no history
no history
—
The full dated event log ships as pricing-history.json, which is named in the publication plan but not yet in the repository.
Where the two studies meet
Quality and speed are close to uncorrelated. The fastest provider ranks last on retrieval; the best retriever is fifth of seven on speed.
Quality and disclosure are close to uncorrelated too. Brave places second on retrieval at 90.0% and scores 1 on compliance and 1 on documentation. Tavily has the strongest disclosure in the field and places sixth of seven on retrieval.
The best-disclosed vendors are not the most independent. Tavily leads compliance and procurement while operating no general index of its own. Brave and Exa own their indexes and disclose least and second-least on compliance respectively.
Nothing here supports a single ranking. A buyer optimising for retrieval, for continuity, or for passing a security review would choose three different vendors from this set of seven.
What we could not determine
Named gaps, not hedges.
Whether any vendor's compliance controls function.
We checked whether reports exist, not what they contain.
Index provenance beyond vendor statements.
Cross-provider result overlap would settle this from data we already hold. Not yet run.
Freshness.
SimpleQA answers are static by design. Providers are known to diverge sharply between static-fact and time-sensitive performance, and nothing here speaks to that.
Whether SimpleQA appears in any provider's own tuning or evaluation set.
It is public and well known. We cannot detect this.
Serper's true price.
Its public pricing page was not reachable and its credit packs sit behind signup.
Non-English and non-US performance.
Untested.
Data and reproduction
Everything published under CC BY 4.0. Attribution required; commercial use, redistribution and tabular reproduction permitted.
File
Contents
Status
manifest.json
Index of all artifacts, versions, dates, schema references
Per-call timings, both passes — 7,000 concurrent and 1,050 sequential
published
pricing-history.json
The 24-month event log
not yet written
provider-config.json
Literal API parameters per provider
not yet written
schema/
JSON Schema for each file
not yet written
Every provider response and intermediate judgement is stored, so scoring rules can be changed and the study re-scored offline without re-querying any API.
Corrections. Logged with the artifact that prompted them. Original values stay visible. We do not change a score in exchange for anything, including access, data, or advertising.
Publication blockers recorded in scorecard.json
These are unresolved at the verification date and are reproduced here rather than held back.
Serper's price_stability and compliance zeros rest on absence of evidence and third-party reporting. Both are load-bearing and must be verified directly.
Brave's pricing cells must be verified against the live pricing page, and the February 2026 announcement needs its URL and full ISO date, which the conventions require and the record still omits.
Brave's partial grandfathering — existing free-plan subscribers at 2,000 queries/month reportedly retaining access — rests on a third-party post and must be checked against Brave's own announcement. It is the evidence behind the anchor_gap on that cell.
Exa's $5 to $7 transition date is not established. Re-derive it from Internet Archive captures of exa.ai/pricing between the 2025 $5 citation and 2026-03-03. Until then the increase is plotted as an undated span.
Exa's 2026-07-14 reductions and the ACU unit change are tracker-sourced. Confirm from a primary source before the ACU cell moves from unresolved to substantiated.
You.com's 2025-10 launch rates are tracker and Archive sourced and must be verified. Section 12 forbids establishing a change from a tracker.
You.com's price_stability carries a conditional: confirm whether the pricing page and the Search API docs disagree on the 100-calls-per-day free quota. A confirmed discrepancy moves the score from 4 to 3.
All compliance cells graded vendor_claimed with verification_status pending must be re-attempted against the live trust centers before the compliance dimension is published as scored.
Tavily's supply_independence rests on a competitor's comparison page. Requires corroboration or downgrade to not_disclosed.
Company founding dates are TODO for Exa, Tavily, You.com and Brave; the age normalisation rule cannot be applied without them.
Right of reply has not been conducted.
Version
unresolved
Reliability verified
2026-09-03
Valid until
2026-12-02
Published
unresolved
Attribution
unresolved
Private evaluation
Run the same test on your workflow.
Bring your requirements and your shortlist. We'll run one free evaluation and show you the evidence behind it.
