02 · CONCEPTS

Credits & pricing

Pay as you go with credits, or choose an API plan for monthly credits and unlimited search. The prices below are the same either way. Every call discloses its exact charge in the response, and misses are free unless a price below says otherwise.

1

The price of everything

VerbPriceCharged when
find_leads0.25 cr / lead · 3 cr minimumper lead actually returned, capped at your limit; at least 3 cr when any lead is returned; zero results free; covered by unlimited search on an API plan
find_companies25 cr hiring · 27 cr lookalike · 28 cr technologyper unique canonical company returned; a plain-language query is priced as the mode it becomes; zero results free
find_companies — lookalikes narrowed by country or size33 cr / company kepteach lookalike filled with size, country and industry; companies that don't fit your country_codes or company_headcount are dropped and free
find_companies — firmographic and local modes2 cr firmographic company · 3 cr local businessper unique canonical company returned (local: businesses with a website); zero results free; firmographic companies are covered by unlimited search on an API plan
find_companies — eventsper proven company · comingcoming, not yet available: companies that had something happen, each with a quote from its source, its date and its own website; companies that can't be proven, and empty searches, are free
enrich_company13 cr / dataset hitprofile, technologies, funding or relationships; empty/unavailable free
enrich_company brief10 cr fast · 40 cr deeponly for a schema-valid evidenced brief
find_posts1.1 cr / relevant post · Reddit and Facebook 2 cronly posts returned by literal-v1; intent sorting is billed separately at llm token prices
find_posts leads3 cr / lead with a signallist mode: a lead's own post, comment, reaction or company post; leads with none are free
find_posts engagers0.5 cr / person · Reddit 2 cr / personengagers mode: each person who reacted, commented, replied or reposted on up to 5 posts; a post with nobody is free
find_posts engagers scan0.4 cr / person readonly with author_filter: each person read before filtering, on top of the per-person price; LinkedIn and X only
find_posts engagers full-read scan0.4 cr / person readonly when reading everyone (read_all) with author_filter: each person read before filtering, on top of the per-person price; LinkedIn and X only
find_posts scoped1 cr base + 0.5 cr / postauthor or company search: the base applies even with zero posts
enrich6 cr / hit · 20 cr / searched mobileonly when a contact is found; cache repeats included, except a crew member's run, which reads an answer its workspace already holds free. A mobile from the separate mobile search (asked for with want: ["mobile_phone"]) is 20 cr, only when found; a mobile that comes with the email is included in the 6
find_leads premium1.6 cr / lead · offer onlyoffered when the standard sources come back short; runs only when you accept the offer (within the key's premium allowance, 0 until an owner raises it) or consent to premium within your budget; its own charge; people already in your answer, and misses, free
enrich premium mobile20 cr · 50 cr / mobile · offer onlycoming, not yet available: only when you accept an offer for a premium mobile source; misses free
verify_email2 cr / callwhen the chain produces an answer
web_fetch0.4 cr / pageonly when a public page is returned; misses and failures free; workspace release required
web_search2.5 cr / searchonly when search results are returned; misses and failures free; workspace release required
research_company9 cr / research · +2 cr / page (deep)once per completed research with supported claims; deep pages only when read; misses and failed rewrites free
prospect_picturecomponents at their verb prices + 3 cr assembly · research from the shared library at 40%each completed component once, assembly only when data is returned; when the company was researched at the same depth in the last 14 days, the research component is served from the shared library at 40% of its fresh price (3.6 cr standard); the exact maximum for your inputs comes from dry_run
find_intent11 cr / company hiring now + 13 cr / paid checkhiring first: a hiring company's latest funding round comes with it and empty hiring checks are free; a paid check (funding on a company not hiring, news or job changes) is charged whatever it finds, only when you ask for it
check_companies5 cr / company answereda yes/no question answered from each company's own site: a yes comes with the quote that states it; unclear answers and unreadable sites are free; a crew member's run repeats the same company and question free while the answer is fresh
setup_sending$15 / domain / year + $7.50 / mailbox / monthcharged to your card, not credits: each domain's first year at purchase, then yearly on your renewal date before its anniversary until released; upkeep monthly per active mailbox
create_campaignfree—
add_leadsfree—
launch_campaignfree—
get_replies · get_reply · set_lead_status · reply_to_thread · hand_off · my_tray · finish_items · unfinished_purchases · goal_progress · assign_lane · ask_colleague · answer_colleague · team_messages · usage · statefree—
brain_read · brain_search · brain_write · account_filefreecoming, not yet available: your team's playbook and the account file of each company and person it works on
crew quality evalper eval · comingcoming, not yet available: after a team's batch reaches you, at most once a day, a judge scores up to 20 of its people and their emails against your playbook; a judge that fails is free
llm — claude-sonnet-50.9 cr / 1k tokensper token, streamed calls reconciled to the exact count
llm — claude-haiku-4.50.45 cr / 1k tokens
llm — gpt-5-mini0.2 cr / 1k tokens
llm — claude-sonnet-5.50.9 cr / 1k tokensavailable on /v1/llm now; for agents that run in Layrcake when hosted agents open; per token, as above
llm — claude-opus-5.51.8 cr / 1k tokensavailable on /v1/llm now; for agents that run in Layrcake when hosted agents open; per token, as above
llm — Claude models by token typeper 1k tokens, input / cached / cache write / output: Sonnet 5.5 and Sonnet 5 0.7 / 0.035 / 0.86 / 3.45 · Opus 5.5 1.4 / 0.07 / 1.72 / 6.9 · Haiku 4.5 0.35 / 0.035 / 0.43 / 1.72Claude models are charged by token type, with thinking counted as output; the blended rates above are the fallback for a model without all four rates, and gpt-5-mini stays blended

Every unscoped miss settles at zero. The fixed outcome fees are the 1-credit base on author/company-scoped find_posts, because the author or company search itself was performed, and the 13-credit paid checks on find_intent, which run only when you ask for them.

The find_leads minimum. A find_leads search that returns at least one lead costs at least 3 credits, because the data source bills for every result page that holds anyone. So 1 to 12 leads cost 3 credits, and from 12 leads up a search costs 0.25 credits a lead. A search that returns nobody is still free.

The shared research library. Public facts from completed research — a dated product launch, a funding round, a partnership, each with its source link — are kept in a shared library that never records who asked. When a prospect_picture needs research on a company that anyone researched at the same depth in the last 14 days, it reuses those facts instead of searching again, and that component costs 40% of its fresh price: 3.6 credits instead of 9 for standard; for deep, 40% of the research plus the pages that entry read, counting no more pages than your picture's budget allows. Whether the library covers a company is decided when the picture runs, so the quote and dry_run show the fresh price as the maximum and you are charged the lower amount. research_company itself never uses the library: it runs fresh, or, when your workspace researched the same company at the same depth in the last 7 days, answers from your own workspace's cache at the research price (6 credits), with no page credits.

2

Holds and exact settlement

When a call starts, its worst-case cost is held against your balance (for find_leads, never less than its 3-credit minimum). When it finishes, the hold settles to the exact amount — never more than quoted, usually less. Interrupted calls are recovered automatically; ambiguous outcomes are reconciled before any charge stands. You can watch holds live in GET /v1/usage as credits_reserved.

3

Sending capacity is a lease

Domains and mailboxes have real recurring vendor costs, so capacity carries disclosed monthly upkeep: $7.50 per active mailbox per month, charged automatically to your card on file (domains are $15 each per year, billed on your renewal date before each anniversary until you release them). The quote you approve shows both numbers before you spend anything. Cancel anytime; upkeep stops at the end of the paid month, and a released domain is not renewed.

If an upkeep charge fails, the mailbox enters a 7-day grace window before suspension — nothing is ever silently torn down.

4

Top-ups & the free tier

New workspaces start with 300 free credits, no card: enough for many full test loops (a find 10 → enrich 3 → verify 3 → draft run costs about 27). Then pay as you go: top up by card in the console (minimum $20). Credits land within seconds of payment, and top-up credits never expire. 1 credit = $0.01.

5

API plans

A plan grants credits every month and adds unlimited people and company search under a fair-use allowance. Verb prices are the same as on pay as you go. Agents and keys carry no fee on any plan.

PlanPriceCredits a monthSearch, each monthSearch, every 5 hours
Starter$99 / month · $990 / year10,00025,000 results2,500 results
Growth$299 / month · $2,990 / year30,000100,000 results10,000 results
Scale$799 / month · $7,990 / year85,000250,000 results25,000 results
Enterprisefrom $1,999 / month225,000+500,000 results50,000 results

What search covers. find_leads (not premium) and firmographic find_companies (by industry, size and location). People and company results draw on one allowance. Past the 5-hour allowance or the monthly ceiling, search is charged in credits at the prices above; nothing is blocked. Every covered call reports layrcake.plan_covered_units, and a search call made under a plan reports layrcake.window_resets_at. The full rules: layrcake.dev/fair-use.

Credits. Plan credits roll over for 3 months, then expire; top-up credits never expire. Credits that expire soonest are spent first. An annual plan is 12 months for the price of 10, charged once a year, and grants its credits month by month.

Changes. Buy or upgrade in the console. An upgrade applies now and charges the price difference for the days left in the period, with the extra credits and search in proportion. A downgrade or cancellation applies at the end of the paid period; the current period is not refunded. If a renewal payment fails, the plan stays in place for 7 days of grace, then the workspace continues on pay as you go.