fshy

Home / Inference

Inception Labs (Mercury API)

docs.inceptionlabs.ai/get-startedInference endpoint, signup token grantinference/inception-labsJSONcomparereport an errorprovider JSON

Free allowance
100 million free tokens assigned to each new account, no payment details required; free-tier limits are 1,000 API requests, 1,000,000 input tokens and 100,000 output tokens per minute
Verdict
triala time-boxed or one-time trial, not a recurring allowance
Cost
free tier
Identity
emailno card, no phone
Binding limit
volume-quota100 million free tokens per new account; billing information is requested when they run low
Evidence
documented5 sources; a first-party page was read and the sentence quoted
Freshness
stale on 2026-11-29checked 2026-10-29 (live); stale on 2026-11-29
Free tier100 million free tokens assigned to each new account, no payment details required; free-tier limits are 1,000 API requests, 1,000,000 input tokens and 100,000 output tokens per minute
unit tokens; metered on request-count; the quick-start page words the grant as 'initially assigned' to each new user and points to the billing page when the tokens run low; no monthly renewal is stated
Pricenone stated
Limits
  • volume-quota (binding) 100 million free tokens per new account; billing information is requested when they run low source
  • request-rate Free Tier per minute: 1,000 API requests, 1,000,000 input tokens, 100,000 output tokens (Pay As You Go: 3,000 requests, 3,000,000 input, 300,000 output) source
Bindingvolume-quota the 100 million token grant runs out before the per-minute ceiling is a constraint
Identityemail no payment details required, quoted from the quick-start page; the sign-up form itself was not captured, so any further check on it is not established
Egressnot published
Interfacerest
Toolingany OpenAI-compatible client against https://api.inceptionlabs.ai/v1, or the vendor SDKs (pip install inceptionai, npm install inceptionai)
Composes witha large one-off token grant with a high per-minute ceiling, suited to a burst of batch calls before any paid plan; the grant, not the rate, is the scarce link
Reliabilitya one-off grant rather than a renewing allowance: once the 100 million tokens are spent the account needs billing information
Notesthe model on the quick-start page is mercury-3; the grant is worded as an initial assignment, so the verdict is trial-only until a renewal is shown.
Evidence notedocumented the grant and the per-minute limits are quoted from the vendor's quick-start and rate-limit pages; no call was made. Renewal behaviour and the sign-up form are not established.
Parsed numbers
  • request-rate 3000 requests
Evidence 5
datesourcewhat it proved
2026-10-28docs.inceptionlabs.ai/get-startedeach new account gets 100 million free tokens and no payment details are required
2026-10-28docs.inceptionlabs.ai/get-startedbilling information is asked for only when the free tokens run low
2026-10-28docs.inceptionlabs.ai/get-started/rate-limitsfree tier allows 1,000 API requests per minute (3,000 on Pay As You Go)
2026-10-28docs.inceptionlabs.ai/get-started/rate-limitsfree tier allows 1,000,000 input tokens per minute
2026-10-28docs.inceptionlabs.ai/get-started/rate-limitsfree tier allows 100,000 output tokens per minute
Unknown values 3
  • $.egress.metered egress metering was not published
  • $.per_object_cap.stated the field does not apply to this kind of offer
  • $.retention.stated the field does not apply to this kind of offer