# How usage is counted Reference · Usage Search is free. Credits pay for the rest: the rerank, deep search, the tools and drafting. Those calls cost different amounts because they are different work. A prior-art search runs a rerank over the corpus; reading an examiner's citations is one indexed lookup. Charging both the same would mean either pricing the cheap calls for the expensive one, or letting a client loop the expensive one at the cheap price. The same tariff applies whether you call the REST endpoint or the MCP tool in front of it, because the tool proxies to the endpoint with your own key. Plans and prices are on the [pricing page](https://antecedint.com/pricing). Sign in and your free key is issued on the spot. ## What each call costs Four bands, so the cost of a call is something you can hold in your head rather than a per-endpoint tariff nobody reads. | Call | What it does | Credits | |---|---|---| | Prior art searchsearch_prior_art | Two-stage retrieval over 10.5M patents: vector recall, then a rerank against your claim text. Every hit carries the passage it matched on. | 10 | | Claim term support mapmap_claim_support | Grounds every claim term in the specification, and for a CIP or continuation in the parent specs, which is what decides whether a term is new matter. | 10 | | Semantic patent-space locatelocate_in_patent_space | Places an idea in the corpus and returns its neighbours, CPC sections and art units. | 3 | | Patent whitespace opportunitiesfind_whitespace | Regions of the space with thin coverage, filtered by section and opportunity. | 3 | | Trace the citation webtrace_citations | Walk the citation web out from a case: the opinions that cite it and the opinions that cite those (direction in), or what it cites and what those cite (direction out), at most two levels and at most 25 edges per node, most negative treatment first. | 3 | | Patent claim textget_patent_claims | Full claim text and abstracts for up to 25 US applications per call. | 1 | | Cited prior art mapget_patent_citations | What an examiner actually cited against an application, examiner citations listed first. | 1 | | Patent neighborhoodget_patent_neighborhood | What sits closest to a given filing, by embedding rather than by classification. | 1 | | Antecedent basis checkcheck_antecedent_basis | §112(b) antecedent-basis defects in a draft or filed claim set. Deterministic, no model call. | 1 | | Art unit profileget_art_unit_profile | One USPTO art unit: what it examines, its examiners, and its allowance and rejection rates. | 1 | | Examiner searchfind_examiner | USPTO examiners by surname or art unit. | 1 | | Examiner statisticsget_examiner_profile | Rejection, allowance, interview and timing statistics for one examiner. | 1 | | CPC code lookuplookup_cpc_code | What a CPC symbol covers, where it sits in the scheme, and which art units examine it. | 1 | | Patent CPC classificationget_patent_classification | The CPC symbols assigned to a published application or granted patent. | 1 | | MPEP sectionget_mpep_section | One MPEP section, with its rules and examiner actions separated from the prose. | 1 | | EPO Guidelines sectionget_epo_guideline | One section of the EPO Guidelines for Examination, in the same shape as an MPEP section. | 1 | | Guidance Cross-Referencesget_guidance_graph | What a guidance section points at and what points at it, one hop in each direction. | 1 | | Office action form paragraphget_form_paragraph | The literal form paragraph an examiner pasted into an office action, by its code. | 1 | | Form paragraphs by rejection typefind_form_paragraphs | Form paragraphs filtered by rejection type or by the MPEP section they rest on. | 1 | | USPTO design search codeget_design_code | One USPTO design search code and what it covers. | 1 | | Read a caseread_case | Read one case from the citation web: its record (court, level, date, author, vote, publication, subject, the decision it reviewed), its good-law status with the dated events behind it, its subsequent history (later opinions reporting it reversed, affirmed, overruled or cert denied, each with the sentence), the overrulings naming it or made by it, and how many opinions cite it and it cites, by treatment. | 1 | | Who cites a caselist_citing_cases | List the opinions that cite a case, each with how it treated the case: treatment (the most negative label among its mentions, with label_counts for all of them), stance, depth (relied on, discussed, passing), whether it quoted, the pin pages, the citing signal and the sentence. | 1 | | What a case relies onlist_cited_authorities | List the cases one opinion cites, each with how the opinion treated it (treatment, label_counts, stance, depth, quoted, pin pages, signal, sentence). | 1 | | Check good lawcheck_good_law | Whether a case was displaced (overruled, abrogated, reversed, superseded), as of a date: the status (displaced, partially_displaced, standing or unverified), the dated events behind it with the deciding case, the overruling edges with the sentence each was read from, and the history reports that displaced it. | 1 | Drawings cost 2. Map tiles, manifests, and anything not listed cost 1. An endpoint that appears without a price in this table costs 1, which is the safe direction: a new cheap route billed at 1 is right, and a new expensive one shows up as our problem rather than as your outage. ## What each plan covers The same allowance read two ways. The left column is what you plan against; the right is what the meter counts. | Plan | Covers | Credits / mo | |---|---|---| | Evaluate | 200 free API searches a day, and 500 credits a month for tools and drafting | 500 | | Launch | 2,000 free API searches a day, and 2,500 credits a month for tools, serving to twice that | 2,500 | Searches and credits are counted apart. A search with ask, without the rerank, spends no credits; the credits go to the tools, the rerank and drafting. Over MCP on your own key, search has no daily limit. ## Ceilings and headroom Three limits apply, and each does a different job. All of them count in UTC: the daily count restarts at 00:00 UTC, and the monthly allowance at 00:00 UTC on the 1st. - **Monthly credits**: The allowance your plan includes. What happens once it is spent depends on the plan.Launch keeps serving past the allowance, up to twice it. Calls in that range are flagged with `X-Antecedint-Plan-Overage` and are not charged. Past twice the allowance, calls are refused until the month resets. Free keeps a daily floor: once the month's credits are spent, 20 credits a day keep working, counted against that UTC day's spend, until the allowance comes back on the 1st. Enterprise limits are set by the agreement. Use past the allowance is forgiven when the month resets. Nothing carries into the next month, and nothing is billed for it after the fact. - **Purchased credits**: Credits bought as a top-up, or committed under an agreement. They do not reset and do not expire, and they are spent only once the monthly allowance is gone, so covering a busy month in advance is not a bet on that month. The balance never goes below zero. - **Requests per day**: A burst guard, so a runaway loop cannot spend a month in an hour. It protects your allowance as well as our capacity. It restarts at 00:00 UTC, and your [account page](https://antecedint.com/account) shows today's count against it. A key that belongs to a firm spends the firm's pool: the monthly credits, the purchased balance and the ceilings above are the firm's, shared by every key in it. ## What counts Tool calls and REST calls spend credits. The MCP protocol messages around them do not: `initialize`, `tools/list`, `ping` and notifications cost nothing, so a client that reconnects often is not charged for it. The front door's lookups (`get_status`, `list_tools`, `describe_tool` and the other list and read tools) are free as well. So is reporting a data error with submit_data_feedback, up to 100 reports a day; a person reviews each one, and an accepted report earns 10, 50 or 200 credits. Runs made on the site count against your key like any other call. That covers the Run buttons in these docs, the sandbox, and the search on the home page when you are signed in: each runs on a one-hour token bound to your key and spends from its allowance. ## Reading it from code Every authenticated response says where you stand, so there is no dashboard to poll. `GET /api/v1/usage` returns the same figures as one JSON document. RateLimit · RateLimit-Policy The daily request limit, in the IETF RateLimit header form: the limit, what remains today, and seconds until the 00:00 UTC reset. X-Antecedint-Credits The monthly credit allowance and what remains of it. X-Antecedint-Plan-Overage: true Present when the call was served from the headroom past your monthly allowance. It is not an error and the response is normal. Treat it as the cue to look at your plan size, not as something to retry or back off from. 429 · USAGE_LIMIT_EXCEEDED A limit was reached. The body reports the daily and monthly figures the decision used, and `credits.required` says what the refused call would have cost, so you can tell a large call hitting a small remainder from an allowance that is spent. `Retry-After` gives the seconds until the limit that refused the call resets. 503 · FREE_TIER_CAPACITY The free tier's shared capacity is full for the moment. It has nothing to do with your allowance: it is a shared pool, and `Retry-After` says when to come back. Paid plans are not subject to it. ## Reading it from an MCP client An agent sees its usage without a separate call. Every tool result ends with a one-line usage summary (what the call cost, what is left this month, and when it resets, in UTC), and carries the same figures as structured data in `_meta`. `get_status` returns the full picture on demand, and costs nothing. A tool call refused on quota comes back as a JSON-RPC error with code `-32003` and `USAGE_LIMIT_EXCEEDED` in its data, rather than as a tool result, so a client can tell a spent allowance from a failed search. ## Throughput Separate from credits, and the ceiling a busy workload usually meets first, including free search over MCP. Rate limits are a property of your key rather than of the deployment, so they move with your plan. | Plan | Throughput | |---|---| | Evaluate | 60 messages/min · 15 tool calls/min | | Launch | 300 messages/min · 60 tool calls/min | | Enterprise | By agreement, with dedicated rerank capacity | A JSON-RPC batch counts as one request for rate limiting but spends the credits of every tool call inside it, so batching is a way to save round trips rather than credits.