Nimble runs two price lists on one page. The self-serve APIs are priced per request, from $1 per 1,000 up to $25.00. The managed Data Services tiers start at $2,500 a month billed annually. Both are published, which is more than most vendors pitched at this end of the market manage. What is not published is narrower and worth naming exactly. Every figure below was read from their pricing page on 9 September 2026.
We sell a competing scraper API, so weigh our reading accordingly. The rates are theirs and the division is ours.
The short version
Extract, Crawl and Map cost $1 per 1,000 requests, Search costs $1.1, and Extract Template costs $3. The Agent API costs $0.025 per request on its low setting, which is $25.00 per 1,000, and only that one setting is priced. Managed Data Services start at $2,500 a month billed annually, which is $7.14 per thousand page credits and a $30,000 annual commitment.
What Nimble publishes on the API side
Six rates, read from the API tab of the pricing page on 9 September 2026. The final column is our own conversion to a common unit.
Nimble API rates
| Product | Published as | Per 1,000 requests |
|---|---|---|
| Free tier | 5,000 requests per month | free |
| Extract, Crawl, Map API | $1 per 1k requests | $1.00 |
| Search API | $1.1 per 1k requests | $1.10 |
| Extract Template API | $3 per 1k requests | $3.00 |
| Agent API, low effort | $0.025 per request | $25.00 |
| Media API | $2 per GB | priced by volume, not requests |
| Custom | talk to sales | not published |
nimbleway.com pricing page, API tab, read 9 September 2026
Four of those rates sit within a factor of three of each other. The fifth sits twenty-five times above the cheapest, on the same list.
The Agent API is a different kind of product
The gap between $1.00 and $25.00 per thousand is not a markup on the same work. Their page describes the Agent API as automating research, enrichment and dataset creation workflows.
It carries an effort setting that the buyer chooses. The page lists low, medium, high, extra high and custom as the options.
Only the low setting has a published price. So the cheapest version of their dearest API is $25.00 per thousand. The rest of that ladder is not on the page.
That is the largest gap in this pricing page, and it sits on their headline product. A buyer sizing an agent workload cannot do it from public figures.
What Nimble publishes on the managed side
The second tab prices a managed service rather than an API. These are people and pipelines rather than endpoints.
Nimble Data Services tiers
| Tier | Per month | Page credits | Concurrent agents | Storage | Per 1,000 credits |
|---|---|---|---|---|---|
| Startup | $2,500 | 350,000 | 5 | 7 days | $7.14 |
| Scale | $7,000 | 1,200,000 | 10 | 30 days | $5.83 |
| Professional | $15,000 | 3,000,000 | 20 | 90 days | $5.00 |
| Enterprise | custom | custom | unlimited | custom | not published |
nimbleway.com pricing page, Data Services tab, read 9 September 2026, with our own division
Every tier is billed annually, so the entry point is a $30,000 commitment. Every tier also says book a demo rather than offering a checkout.
The per-credit rate improves as you climb, from $7.14 to $5.00 per thousand. That is a 30 percent cut across a sixfold increase in spend.
The two lists are not the same product
Read across the two tables and the arithmetic looks strange. Extract costs $1.00 per thousand on one tab, and a page credit costs $7.14 on the other.
That is not a like-for-like comparison and we are not presenting it as one. The managed tiers include concurrent agents, data storage, custom ETL and an integration built by their team.
The useful point is what the gap represents. Choosing between the two tabs means deciding whether the management layer is worth several times the raw rate.
That question has a real answer for each team, and no pricing page can give it. What a pricing page can do is state both numbers, which this one does.
Where the public rates land
Set the published Nimble rates beside the other conversions from this batch. Every figure was read in the same week and restated in the same unit.
Cost per 1,000 requests across the market
| Where you buy it | What it is | Per 1,000 |
|---|---|---|
| Zyte, $500 commitment | HTTP, easiest tier | $0.06 |
| ScrapingBee, Startup | plain request | $0.10 |
| Scrapfly, Pro | datacenter request | $0.10 |
| ScraperAPI, Business | standard page | $0.10 |
| DataForSEO, standard queue | Google SERP | $0.60 |
| Massive, Web Render API | rendered page | from $0.80 |
| Firecrawl, Standard | page scrape | $0.83 |
| Nimble, Extract | page contents | $1.00 |
| Nimble, Search | search request | $1.10 |
| Nimble, Extract Template | structured domain data | $3.00 |
| Nimble, Data Services Startup | managed page credit | $7.14 |
| Nimble, Agent API | low effort request | $25.00 |
| ScraperAPI, Hobby | ultra premium rendered | $36.75 |
| HProxy scraper API | successful call | published on our scrapers page |
eight vendor pricing pages and our own arithmetic, all read 9 September 2026
Our own unit is one successful call from a balance that does not expire. There is no plan and no effort setting to choose. Nimble's Extract rate sits mid-table, and its Agent API sits near the top.
The free tier is the most generous in this batch
Five thousand requests a month, across every API, with no card required and cancel anytime. That is what their page states.
Priced at their own Extract rate, those requests are worth about $5 a month. Priced at their Agent API rate, they are worth $125.
More importantly, it is not a one-off trial. A free tier that renews monthly lets a team build an integration properly. Nobody has to rush a benchmark before a clock runs out.
We compared free tiers across this batch and most give a single grant of credits. This one and Scrapfly's both continue, and Nimble's is the larger of the two by request count.
Who this pricing suits
Nimble suits teams whose work maps onto Extract, Crawl or Search at a dollar per thousand. That rate is competitive without being the cheapest in the market.
It suits agent workloads only after a conversation. The one published effort level is $25.00 per thousand, and the rest is not shown.
It suits buyers who want a managed pipeline and can commit $30,000 a year. That is a serious floor and the page states it plainly rather than hiding it behind a form.
It suits price-led small buyers poorly on everything except the free tier. Three vendors in this batch publish plain page rates ten times below Nimble's. We set those out in the Zyte, ScrapingBee and Scrapfly teardowns.
What we could not verify
The Agent API is priced only at its low effort setting. Medium, high, extra high and custom appear as options with no rate attached. We did not estimate one.
The Search API card shows lite and fast modes at $1.1 per thousand. It also mentions a deep mode that includes page contents. No separate price for deep mode appeared, so we recorded the published figure only.
Everything under the Custom heading is quoted rather than published. That covers volume bundling, throughput SLAs, multi-year price protection, custom concurrency, higher rate limits and PII masking.
We could not confirm whether a managed page credit is the same unit as an Extract request. We therefore report both rates and decline to convert between them.
No statement appeared about whether unused page credits roll over on the managed tiers. On an annual commitment that term matters, and we make no claim about it.
Prices move. Every figure here was read on 9 September 2026, so check the live page before committing.
Sources
- Nimble, pricing page, API tab, read 9 September 2026. The free tier of 5,000 monthly requests, and the rates for the Agent, Search, Extract, Extract Template and Media APIs.
- Nimble, pricing page, API tab, read 9 September 2026. The Agent API effort options listed as low, medium, high, extra high and custom.
- Nimble, pricing page, Data Services tab, read 9 September 2026. The Startup, Scale and Professional tiers with their monthly prices, page credits, concurrent agents and storage periods, all marked billed annually.
- Nimble, pricing page, read 9 September 2026. The Custom heading and the features listed under it as quoted rather than published.
- Zyte, ScrapingBee, Scrapfly, ScraperAPI, Firecrawl, DataForSEO and Massive pricing pages, all read 9 September 2026, for the market comparison table.
- Our own arithmetic on those published figures, 9 September 2026, shown in every table above.