We-Connect Overview
Two products from one company: a LinkedIn app you pay for by the seat, and an API called We-Link you pay for by the credit.
We-Connect is a cloud LinkedIn automation app priced per seat, and a seat is one LinkedIn account you connect: $49 or $59 a month on annual billing. The same company sells We-Link, an asynchronous LinkedIn API with an MCP server, priced in prepaid credits from $39 for 4,000. Which one fits comes down to whether you want campaigns run for you or actions that your own code or agent calls.
On this page
- What We-Connect and We-Link are
- How an account connects and an action runs
- What We-Connect costs
- What We-Link costs and what a credit buys
- The limits each product publishes
- The API and MCP surface
- What the vendor says about account safety
- The shape of job each one fits
- Alternatives to We-Connect and We-Link
- FAQ
- Sources
What We-Connect and We-Link are
We-Connect has sold LinkedIn automation since 2018, and the summary file it publishes for answer engines calls it “a cloud-based LinkedIn intelligence and automation platform for sales teams, recruiters, agencies, and founders”. We-Link is the newer product, a LinkedIn API whose changelog starts at v1.0.0 in December 2025. The We-Link terms settle who stands behind both: “We-Link API is a service owned and operated by We-Connect Systems LLC.”
| Dimension | We-Connect (we-connect.io) | We-Link (we-link.ai) |
|---|---|---|
| What it is | A cloud app that builds and runs LinkedIn campaigns | An asynchronous LinkedIn API with a hosted MCP server |
| Account model | “A seat is one LinkedIn account connected to We-Connect” | “Unlimited accounts, no per-seat fee” |
| Price unit | Per seat per month | Prepaid credits with a validity window |
| Entry price | $49 a seat billed annually, $69 month to month | $39 for 4,000 credits, valid one month |
| Free entry | 14-day trial, no card | 100 credits for 14 days, no card |
| API | Ten read endpoints and one write, none of which sends | 73 documented operations in 15 groups |
| MCP server | Read-only, live on all plans. Campaign building marked coming soon | 11 live tools, labelled beta in the docs |
| Account types | Standard, Premium, Sales Navigator, Recruiter, Recruiter Lite | Classic, Premium, Sales Navigator, Recruiter Lite, Recruiter Corporate |
We-Connect is the finished tool: you connect an account, build a campaign on a canvas, and the vendor runs it. We-Link hands the same kind of LinkedIn actions to your code, one asynchronous call at a time, and leaves the campaign logic to you. The vendor’s developer page says the We-Connect API, by contrast, “is a data layer for the We-Connect campaign platform”.
- The LinkedIn Lead Engine. A drag-and-drop canvas with branching, separate paths for connected and unconnected contacts and 22 channel action nodes, on every plan and still marked beta.
- Over 20 LinkedIn actions. Connection requests, messages, InMail, follows, likes, endorsements, event invitations and greetings on birthdays, new roles and job anniversaries.
- AI in the campaign. Agent nodes score prospects from 0 to 100 and sort post engagement by sentiment, and AI auto-reply runs as a co-pilot or in an autonomous mode that “sends replies automatically without requiring approval”.
- One inbox. LinkedIn, Sales Navigator and Recruiter threads in one place.
How an account connects and an action runs
We-Connect takes the LinkedIn login. In the help center’s words, “Once you log in to We-Connect using your LinkedIn credentials”, you choose a location and get “a residential IP from that location”, set the time zone campaigns run on, and set sleep hours, at least six hours a day, in which “We-Connect pauses all activity”. The network syncs within 24 hours, and campaigns then run on the vendor’s servers whether or not your browser is open.
We-Link has two ways in. One call, POST https://api.we-link.ai/api/v1/login_v2, carries the account’s email and password, an optional twoFASecret from an authenticator app, and a proxy. The visual route opens “a streamed LinkedIn session running in We-Link’s secure browser”, where the owner signs in directly and the credentials never pass through the API. When an authenticator secret is given, “We-Link stores it securely” and uses it to sign the account back in when a session runs out.
Each account needs exactly one proxy before the login is accepted: the internal option, a dedicated IP in one of 69 countries at 410 credits a month, or a proxy of your own, validated first. Changing the location later allocates a new proxy, which “may result in additional charges”. From there every action is asynchronous.
- A requestId comes back at once. The result arrives at your webhook, signed with an HMAC hash, or you poll
POST /api/v1/get_response. - Execution is sequential and paced. The docs put one request at anywhere from a few seconds to several minutes, and a request that waits too long fails with
QUEUE_TIMEOUT_THRESHOLD_EXCEEDED. - Calls can be scheduled. A
scheduledfield in UTC, at least two hours ahead.
Recruiter accounts carry one more step. The docs warn that signing in through We-Link “may trigger LinkedIn’s credential sharing warning”, which the user has to acknowledge, and the API answers RECRUITER_ACKNOWLEDGEMENT_REQUIRED until they do.
What We-Connect costs
We-Connect prices per seat, and the FAQ defines the seat without hedging: “A seat is one LinkedIn account connected to We-Connect.” Teammates who do not connect a LinkedIn account of their own cost nothing. The pricing page opens on annual billing, the FAQ prints the yearly totals, and the summary file carries the monthly prices.
| Plan | Per seat / month, billed annually | Paid up front per year | Month to month | What the plan adds |
|---|---|---|---|---|
| Growth | $49 | $588 | $69 | Unlimited LinkedIn campaigns, the Lead Engine, AI Assist and auto-reply, contact enrichment |
| Professional | $59 | $708 | $79 | LinkedIn and email in one campaign, native CRM sync, email tracking |
| Scale (marked Soon) | $79 | $948 | $99 | The Watchlist with up to 4 intent agents, multi-sender campaigns, Post Remix |
| Agency | Custom | Custom | Custom | 10 or more seats, a dedicated success manager, custom integrations |
Two of the four rows cannot be bought from the page today: Scale has a Notify me button where a checkout would be, and Agency is a demo booking. The trial runs 14 days with no card. Post Remix, the post scheduler, is printed two ways: an “Introductory Price” of $16 a seat a month billed annually on the pricing page and “$19/seat/month” in the summary file. The Watchlist, which enrolls prospects in campaigns from signals such as profile visits, is marked “Limited time” on the two lower plans and is due to move to Scale.
No volume break is published below Agency, so the arithmetic is linear. Ten seats on Growth come to $490 a month on annual billing, $5,880 paid up front, or $690 a month on monthly billing.
What We-Link costs and what a credit buys
We-Link sells prepaid credits, and every plan, in the pricing page’s words, “lets you connect as many accounts as you like”. There are four plans, each with a validity window, plus 100 free credits for 14 days with no card. The last column is our arithmetic, at the 5 credits the docs charge for most calls.
| Plan | Price | Credits | Per credit | Validity | One 5-credit call |
|---|---|---|---|---|---|
| Starter | $39 | 4,000 | $0.0098 | 1 month | about 4.9 cents |
| Growth | $99 | 12,000 | $0.0083 | 3 months | about 4.1 cents |
| Scale | $249 | 40,000 | $0.0062 | 6 months | about 3.1 cents |
| Pro | $490 | 100,000 | $0.0049 | 12 months | about 2.5 cents |
Every plan carries “All 70+ endpoints”, MCP and REST, Sales Navigator and Recruiter, and auto top-up. A top-up block is billed at the plan’s rate and “expires with your plan”. The arithmetic matters because the vendor’s own pages give four readings of what a credit buys.
- The pricing page counts credits. Starter is “4,000 API credits at $0.0098 / credit”.
- The developer hub counts calls. “Plans start at $39 for 4,000 calls”, and the trial is “100 API calls over 14 days”.
- The rate-limits page charges one credit a call. “Each API call consumes 1 credit from your account balance.”
- The credit-usage page charges most calls five. Of its 82 priced paths, 60 cost 5 credits, among them connect, send a message, view a profile and every search.
On the first three readings $39 buys 4,000 calls. On the credit-usage table it buys about 800 connection requests or messages, and fewer on the internal proxy: one account on Starter then has 3,590 credits left after the proxy’s 410, about 718 actions at 5 credits each.
| Credits per call | What the credit-usage page charges it on |
|---|---|
| 0 | Files, listing accounts, proxy status and switching, message trackers, submitting a login code (14 paths) |
| 3 | Editing or deleting a Recruiter InMail follow-up |
| 5 | 60 paths, among them connect, message, InMail, view a profile, like, follow, endorse, search, and polling a result |
| 8, 12, 14 | Login, own profile, connection status and company details at 8, message threads at 12, the connections list at 14 |
| 410 a month | The internal proxy, per account |
Two lines shape how an integration is written: polling a result through get_response costs 5 credits, so the webhook is the cheaper way to hear back, and the connections list costs 14, the most of any call.
The docs also keep an earlier model. The subscription page still says “Each block includes 100,000 credits” and that credits “are valid for one year”, which is how the We-Link homepage read on 21 September 2026, at “$0.0049 / credit”. The Pro plan keeps that rate and that validity, and the three smaller plans arrived after it.
The limits each product publishes
Both products publish numbers of different kinds. We-Connect publishes a weekly pace it recommends for each action, plus daily ceilings. We-Link publishes a rate limit per account per hour, enforced on its side.
We-Connect’s recommended weekly pace. The usage guidelines say where the numbers sit: “These are We-Connect’s recommended weekly limits. They are intentionally set below LinkedIn’s platform limits to keep your account safe.” A paid account is LinkedIn Premium, Sales Navigator or Recruiter.
| Action | Free LinkedIn account | Paid LinkedIn account |
|---|---|---|
| Connection requests | 100 a week | 200 a week |
| Messages to 1st-degree connections | 500 a week | 700 a week |
| Free InMails to open profiles | 250 a week, counted toward messages | No separate figure |
| Profile visits | 200 a week | 350 a week |
| Follows and likes | 200 a week each | 300 a week each |
| Endorsements and invitation withdrawals | 140 a week each | 200 a week each |
| Event invitations | 100 a week | 250 a week |
| Company Page invitations | 30 a month to your employer’s Page, 50 to other Pages | No separate figure |
Above that pace sit two ceilings, stated in the warm-up article: “In total, We-Connect can take 150 LinkedIn actions per day, excluding messaging. We-Connect can send 200 LinkedIn messages per day.” Inside them, each action type has a global limit you set per account, starting at “40% of its full potential”, and each campaign has a daily action limit shared across everything it does, so a limit of 25 can mean 15 connection requests and 10 profile visits.
- Warm-up, with two defaults on one page. The same article says a new account gets it “turned on automatically by default” and, three steps later, “By default, account warm-up is turned off.” The usage guidelines say warm-up “keeps invites to no more than 15 per day”.
- Randomized volume, with two worked examples. One article varies a limit of 20 to “anywhere between 15 and 25 actions per day”, another a limit of 30 to “25 to 30 requests”. Both call it “within five actions”.
- Sleep hours. At least six hours a day with no syncing and no campaign activity, in the account’s time zone.
- A pause when the account is limited. “If LinkedIn ever limits your account’s activity, we pause your campaigns right away”, and they “resume automatically after one week” unless restarted sooner.
We-Link’s limits are per hour and per account. The global action rate limit is 100 requests an hour and message syncing is 150, each over a rolling 60 minutes, and past either one the API answers 429 Too Many Requests. At the ceiling one account can run 2,400 actions a day, which at 5 credits each is 12,000 credits, a whole Growth plan. Below it the docs recommend a pace and leave it to the caller: one request at a time, 5 to 15 seconds between actions, 2 to 5 minutes between batches. The MCP server adds “rolling daily quotas” and answers QUOTA_EXCEEDED_DAILY, without printing what the quotas are.
The same docs print a table of LinkedIn’s own limits by account type, noting against profile views that the figure is “Not enforced by We-Link”. LinkedIn publishes no numeric invitation cap, and what it does document is on our LinkedIn connection limits page. The We-Connect API has a limit of its own, unrelated to LinkedIn actions: “300 requests per minute per account”.
The API and MCP surface
The developer hub lists four surfaces and says which to pick: “Embedding LinkedIn actions into your product? Use the We-Link API.” An assistant that acts gets the We-Link MCP server, campaign data comes from the We-Connect API, and a read-only We-Connect MCP server answers questions about your outreach, “Read-only today, with campaign building on the way.”
The We-Link REST API. Calls go to https://api.we-link.ai/api/v1/ with x-api-key and x-api-secret headers, and each key can be restricted to listed IP addresses and to a chosen set of actions. The marketing pages say “more than 70 endpoints in 14 groups”. The API reference, counted group by group on 15 October 2026, lists 73 operations under 15 headings, the fifteenth being Realtime Event Tracking.
| Group | Operations | What is in it |
|---|---|---|
| Authentication | 3 | Log in, submit a one-time code, fetch your own profile |
| Account management | 9 | Connected accounts, profile viewers, subscriptions, Recruiter contracts, InMail credit balance |
| Connections and networking | 10 | Send, check and withdraw requests, connections and pending invitations, profile and contact details |
| Messaging | 11 | Send to a profile, thread, group or event member, InMail, read threads, follow-up trackers |
| Search operations | 9 | Regular, saved and history search, posts, groups, events, Sales Navigator lists, Recruiter projects |
| Profile engagement | 5 | View, follow, unfollow, endorse, like a post |
| Company, events, posts | 8 | Company details and follow invitations, event invitations, create a post, read a profile’s posts |
| Greetings | 6 | Birthdays, job changes and work anniversaries, and a greeting for each |
| Files, listener, proxy, status, usage | 12 | Uploads, a realtime event listener, proxy switching and rotation, request status, credit usage |
Recruiter Corporate support arrived in March 2026, per the changelog. Results arrive as webhook events such as CONNECTION_SENT and MESSAGE_REPLY, and n8n has a documented integration.
The We-Link MCP server. It is hosted at https://mcp.we-link.ai/mcp and takes three credentials: the API key, the API secret and the ID of one connected account, because “Each MCP connection uses one account”. In Claude’s custom connector all three go into the connection URL as query parameters, and Cursor’s configuration sends them as headers. The docs page is titled “We-Link MCP Beta”, the developer hub badge reads “Live · Beta”, and the pricing page says “The We-Link MCP server is live.”
| Status on the MCP page | Tools |
|---|---|
| Live, 11 | Search profiles, get profile, send connection request, send message, get pending invites, get message threads, get messages from a thread, my connections, get posts, like post, get post reactions |
| Coming soon, 7 | Get my profile, profile views, connection status, follow or unfollow, disconnect, endorse skills, celebrations |
| Planned, 5 | Post stats, events, groups, subscriptions, Recruiter |
Two of the live tools send (the connection request and the message), under what the MCP page calls the REST API’s own protection: “a dedicated, geo-matched IP, human-like pacing and automatic rate management”. When LinkedIn asks for verification mid-session, the user pastes the one-time code “directly into your MCP client chat”. What changes when an agent drives an account through tools like these is the subject of our LinkedIn MCP server page.
The We-Connect API. A much smaller surface, version 3.0, at https://api-us-1.we-connect.io, with the key sent as an api_key query parameter. Ten of its eleven endpoints read: the account, campaign stats, connections, contacts with an email, tagged contacts, leads, follow-ups, notes, LinkedIn replies and email replies. The eleventh imports contacts into a campaign. Nothing in it sends a message or an invitation.
The We-Connect MCP server, live and read-only. It connects Claude, ChatGPT or another MCP client to the workspace and answers questions about campaign results, the inbox, the network, Watchlist people, restriction risk, daily limits and pending invites. It sends nothing: “It doesn’t send messages or connection requests”, it does not change campaigns or settings, and it never returns message text or contact details. Building campaigns from a sentence is the next step, marked coming soon. The page also says who holds the brakes: “Safety is enforced by We-Connect, not by the assistant.”
What the vendor says about account safety
Both products answer the safety question with controls, and neither site publishes an outcome figure. The controls they name, in the vendor’s wording where there is one:
- A dedicated IP per account. We-Connect lists a “Dedicated country-based IP address”, residential and from the account holder’s home country per its summary file. We-Link says “Every account gets a dedicated, geo-matched IP”, or takes your own proxy.
- Pacing and variation. “Randomized activity timing” and a daily volume that varies within five actions on We-Connect, sequential processing with “human-like random delays” on We-Link.
- Caps you set, and caps the vendor sets. We-Connect’s global and campaign limits, which the FAQ says “are configurable per seat”. We-Link’s cap is its own, 100 actions an hour per account, plus the MCP server’s daily quotas.
- Warm-up, sleep hours and invitation clean-up. All three on We-Connect, including automatic withdrawal of pending connection requests. We-Link publishes a four-week warm-up schedule and leaves following it to the caller.
Both products connect with the account’s LinkedIn login. We-Link adds the streamed sign-in, which keeps the password out of the API call, and keeps the authenticator secret so it can sign the account back in by itself. Its security page lists AES-256 encryption at rest, data stored in the United States and SOC 2 as “On our roadmap”, and its terms put the platform question on the customer: “You are solely responsible for ensuring your use of the Service complies with LinkedIn’s terms of service and policies.”
The shape of job each one fits
We-Connect fits a person or a sales team that wants LinkedIn campaigns built on a canvas and run by the vendor, on accounts they own, with the pace published action by action and the caps theirs to set. Its edges are a per-seat price with no published volume break below Agency, two of four plans not yet on sale, and an API that reads campaign data without acting on LinkedIn.
We-Link fits a developer who wants LinkedIn actions inside their own product, across any number of accounts, billed by use, and who is comfortable working asynchronously: send a call, receive a webhook. Its edges are about planning: what a credit buys depends on which page you read, one account tops out at 100 actions an hour, credits expire with the plan, and Recruiter is on the MCP server’s planned list.
Alternatives to We-Connect and We-Link
GTM API is our product, and it is listed first. The other rows are described from their own pages, read on 14 October 2026.
| Tool | What it is | Price unit | Entry price | MCP server |
|---|---|---|---|---|
| gtm-api.com | LinkedIn REST API and hosted MCP server, one cloud browser per account | Per connected account, calls not metered | Free plan, then from $10/account | Hosted, 250+ typed tools, sends behind a confirmation step |
| Unipile | One API for LinkedIn, WhatsApp, email and calendars | Per connected account | EUR 49 a month for up to 10 accounts | A documentation server for coding agents |
| Linked API | A workflow API, one cloud browser per seat | Per seat, one account each | $49 a seat billed annually, $69 monthly | Hosted, source on GitHub |
| LinkupAPI | One API for LinkedIn, email and WhatsApp | Per connected account | $55 a month for up to 10 accounts | Hosted, OAuth 2.1 or an API key |
| ConnectSafely.ai | API, MCP server and CLI in one plan | Per LinkedIn account | $10 an account a month | Included in the same plan |
| HeyReach | A cloud app that rotates one campaign across several accounts | Per sender, one account each | $79 a sender a month on Growth | A server per workspace |
| Aimfox | A cloud app for LinkedIn campaigns | Per seat, one profile each | $49 a seat a month | Hosted, OAuth |
GTM API covers the We-Link side of the job: LinkedIn actions as typed calls, over REST or the hosted MCP server at mcp.gtm-api.com, on accounts you own. Each connected account runs in its own cloud browser with its own proxy, and the LinkedIn password is never handed over, because sign-in happens inside that browser. Daily budgets per action are checked on the server before every call, and over MCP anything that sends returns a preview and waits for a confirmation. The MCP server exposes 250+ typed tools across 19 toolsets, 190+ of them on the LinkedIn side across 12 toolsets. Recruiter is covered on an account that holds a Recruiter seat: candidate search, the talent inbox, hiring projects and Recruiter InMail. Pricing is per connected account with calls not metered: a free plan with one account and no card, then from $10/account. What a LinkedIn automation API is, and how ten of them price and pace, is on LinkedIn automation API.
Five of the others have an overview of their own here: Unipile, Linked API, ConnectSafely.ai, HeyReach and Aimfox. HeyReach and Aimfox are the two rows that sell what We-Connect sells, a campaign tool priced per connected account.
The first question that sorts the field is the one that already splits We-Connect from We-Link: do you want campaigns run for you, or actions your own code calls? The second is the unit the bill counts, a seat, an account or a credit spent per call.
Frequently Asked Questions
What is We-Connect used for?
Running LinkedIn campaigns from the cloud on accounts you own. You connect an account, build a campaign on a visual canvas, and We-Connect sends the connection requests, messages, InMails, follows and likes on a daily schedule, with AI drafting openers and replies. The Professional plan adds email in the same campaign. No browser extension is involved.
Are We-Connect and We-Link the same company?
Yes. The We-Link terms say We-Link API is a service owned and operated by We-Connect Systems LLC. They are two products with two price units: We-Connect is a campaign app priced per seat, and We-Link is an asynchronous LinkedIn API with an MCP server, priced in prepaid credits. We-Connect also has a small API of its own that reads campaign data.
How much does We-Connect cost?
Per seat, where a seat is one connected LinkedIn account. Growth is $49 a seat a month and Professional $59 on annual billing, paid up front at $588 and $708 a year, or $69 and $79 month to month. Scale, at $79 on annual billing, is marked Soon and cannot be bought yet, and Agency is a custom price for ten or more seats. The trial is 14 days with no card.
How much does the We-Link API cost?
Prepaid credits on four plans, from Starter at $39 for 4,000 credits valid one month to Pro at $490 for 100,000 valid a year, which is $0.0098 down to $0.0049 a credit. The credit-usage page charges 5 credits for most calls, such as a connection request or a message, so one call costs about 4.9 cents on Starter and 2.5 cents on Pro, while the rate-limits page says a call costs 1 credit. New accounts get 100 credits for 14 days with no card.
What are We-Connect’s LinkedIn limits?
We-Connect recommends a weekly pace per action: 100 connection requests a week on a free LinkedIn account and 200 on a paid one, 500 or 700 messages to connections, and 200 or 350 profile visits. Above that, the app tops out at 150 LinkedIn actions a day excluding messaging and 200 messages a day. Each action type has a global limit you set per account, starting at 40% of its maximum.
Does We-Connect have an API?
Two, for two jobs. The We-Connect API reads campaign data (the account, campaign stats, connections, leads, follow-ups, notes, tags and replies), and its one write imports contacts into a campaign. It sends nothing. Sending through an API is We-Link’s job: a REST API with 73 documented operations in 15 groups, asynchronous, with results returned by webhook.
Does We-Link have an MCP server?
Yes, hosted at mcp.we-link.ai and labelled beta in its docs. It takes an API key, an API secret and the ID of one connected LinkedIn account per connection. Eleven tools are live, including searching profiles, sending a connection request and sending a message, with seven more coming soon and five planned, Recruiter among them. We-Connect’s own MCP server is live and read-only, and building campaigns from it is marked coming soon.
What are the alternatives to We-Connect?
It depends which of the two products you are replacing. For LinkedIn actions as API calls, GTM API, our own product, has a REST API and a hosted MCP server on accounts you own, priced per connected account with calls not metered, from $10/account after a free plan. Unipile, Linked API, LinkupAPI and ConnectSafely.ai are account-based APIs too. For a campaign app priced per connected account, HeyReach bills per sender and Aimfox per seat.
Sources & Further Reading
- We-Connect, pricing (plans, seat definition, yearly totals, plan grid, Post Remix) · summary file for answer engines (monthly prices, trial, founding year, account types) · developer hub (the three surfaces, pricing in calls) · LinkedIn MCP (the coming-soon builder), read 14 and 15 October 2026
- We-Connect help center, usage guidelines (weekly table, campaign limit, pause) · warm-up and global limits (daily ceilings, the 40% default, both warm-up defaults) · connecting an account · sleep hours · randomize · API reference (version, endpoints, rate limit), read 15 October 2026
- We-Link, homepage · pricing · MCP server (live, coming-soon and planned tools) · terms (the operator, section 4.3) · security, read 14 and 15 October 2026
- We-Link docs, credit usage · subscription · rate limits · safety and platform limits · what is We-Link · platform restrictions (the warm-up schedule) · authentication · asynchronous processing, read 15 October 2026
- We-Link docs, connecting an account · visual connection · session persistence · proxy configuration · MCP overview · MCP in Claude · MCP in Cursor · API reference (the 73 operations) · webhook events · n8n · changelog, read 15 October 2026
- Alternatives, pricing and MCP pages read 14 October 2026: Unipile pricing, MCP · Linked API pricing, MCP · LinkupAPI pricing, MCP · ConnectSafely.ai pricing, MCP · HeyReach pricing, MCP · Aimfox pricing, MCP
- gtm-api.com, product · pricing · developer documentation: MCP server, smart limits, browsers and proxies, Recruiter, billing and plans
- Related: LinkedIn automation hub · Best LinkedIn API · Best LinkedIn automation tools · LinkedIn API pricing · LinkedIn connection limits · LinkedIn MCP server · Linked API overview · HeyReach overview
LinkedIn actions as typed API calls.
Connect an account server-side, send a connection request or a message as one endpoint, and read the limits before the call goes out. Free plan, then from $10 per connected account at volume, with calls not metered.
Last updated: October 2026 · Definitions and benchmarks verified at time of publication