Pricing

Cost + a 5% platform fee. Nothing else.

Provider cost plus a 5% platform fee on every API call — that's our entire revenue model. No monthly fee, no seat count, no minimum. Cached requests get a 50% discount.

Adding credit carries a separate 5% top-up fee that covers payment processing — shown before you pay, same as anywhere you check out online with a card.

$1 free* · add a card, no chargeTopups never expireRefund within 90 days

Trial · add a card

$1 free*no charge to add a card
A card is required to call the API. Adding one never charges you and grants $1 across the curated trial catalog.
  • $1 in credit across 15 curated models, led by GLM-5.2
  • Kimi K2.7 Code, gpt-oss-120b, BGE-M3, FLUX, Moondream, Aura 2, Whisper Turbo and more
  • Full playground, logs, and analytics
  • $1 expires 30 days after you add a card; topups never expire
Claim $1 on GLM-5.2

Pay as you go

cost + a 5% platform fee
Unlocks everything the moment you add credits.
  • All 1000+ models — Opus 4.7, GPT-5.4, Gemini 3.1 Pro, Kimi K2.7 Code, Llama 4, Flux 1 Schnell, Veo 3.1, MiniMax Hailuo
  • Every modality: text, image, video, music, STT, TTS, vision, embeddings, rerank, classification
  • Sub-account API · nested keys for your end users
  • Cost-attribution tags · hard budget caps · BYOK
  • Semantic + exact cache · logs · usage analytics
  • Bill only successful runs · 50% off cached requests
Add credits

Enterprise

Custom/from $10k/mo committed
Everything in PAYG, plus the primitives only an aggregator can ship.
  • Evals — replay real traffic against alternate models, score against your rubric
  • Guardrails — content safety + PII redaction, one policy every provider
  • Replay + shadow A/B — deterministic re-run, canary new models safely
  • Prompt IDs — versioned prompts server-side with auto prefix caching
  • SSO · SCIM · dedicated endpoint · security posture
  • 99.95% SLA · DPA · SOC 2 · private audit export
  • Direct-provider agreements · named engineer on Slack Connect
Talk to sales
Calculator

What you'd actually pay us.

Plug in your monthly volume. We show you side-by-side cost against the aggregators we replace. Cache savings are modeled separately.

Already paying OpenAI, Anthropic, Google, or OpenRouter? See what your actual bill would look like →

$2.50 in / $10 out
120M
40M
30%
$738.50
5.5% platform fee
$741.00
~3% + platform min
$624.75
− $113.75 vs A
BREAKDOWN · AIGATEWAY
Provider token cost$700.005% platform fee$29.75Cache savings− $110.25Net$624.75
How we compare

Cheaper than OpenRouter.
Primitives none of them have.

Numbers pulled from each provider's public pricing page in April 2026. We update this monthly.

FeatureAIgatewayOpenRouterPortkeyHeliconeRequesty
Platform fee (our only margin)5% per call5.5% on credits$49/mo flat$79/mo+ flatfree
Card top-up fee5% · payment processingin the 5.5%variesvaries—
Cache-hit discount50% off cached requestsnononono
Every modalitytext · image · video · voice · audio · embedtext onlytext onlytext + obsvaries
Sub-account / per-user keysyes, APInoworkspace onlyworkspace onlyno
Cost-attribution tags + hard capsyes, APIbasicyesyesno
BYOKfree · no per-request fee1M free/mo then 5%includedincludedyes
Evals on your trafficEnterprisenorules onlynono
Guardrails (content + PII)Enterprise · one policy, every providernoyesnono
Replay + shadow A/B across modelsEnterprisenonoprompt-level onlyno
Prompt IDs with auto-cachingEnterprisenoworkspace onlynono
Auto Router

Let us pick the model.
Pay even less.

Cost + a 5% platform fee is the floor, not the only option. Set model:"auto" and the router picks the cheapest model in a curated pool that still clears the quality floor — cheaper than the premium model you'd otherwise call, guaranteed, with the baseline acting as a hard cost ceiling. Works across text, image, video, speech, transcription, music, and embeddings.

Every routed call returns headers showing the model that ran and the exact dollars saved versus your baseline. You keep the majority of every dollar the router saves, and you never pay above that baseline. The full pricing mechanics are in the docs fine-print.
See the Auto Router →Read the docs
FAQ

Questions we actually get.

What's the catch on 5%?
There isn't one. 5% per call is our entire platform fee — the only margin we make. We add it to the underlying model cost on every API call. No monthly fees, no seat fees, no per-model surcharges, no markup on tokens, no minimum. You only pay when a request succeeds. (Adding credit carries a separate 5% top-up fee that covers payment processing — disclosed before you pay, like any online card checkout.)
So what does it actually cost me?
Per call: provider cost × 1.05 — debited from your wallet when the request succeeds, and that 5% is the only fee we earn. Adding credit: a one-time 5% top-up fee covering card processing. Worked example: buy $100 of credit → charged $105 → wallet credited $100 → each call debits provider cost + a 5% platform fee from the wallet.
Is there free credit?
Adding a payment method (no charge) gets you $1 in credit spendable across a curated trial catalog — GLM-5.2, Kimi K2.7 Code, gpt-oss-120b, FLUX, Whisper Turbo, Aura 2, BGE-M3 and more. The $1 expires 30 days after you add the card. Top up any amount and the full 1000+ catalog opens at pass-through + a 5% platform fee. Topups never expire.
What counts as a 'successful run'?
A request that returned a usable model response. Fallbacks that succeed bill once, at the model that actually answered. Failed runs and timeouts don't charge.
How do cache hits price?
Exact-match and semantic hits return in under 10ms and get a flat 50% discount on the uncached price. Cache is on by default; override per-request with the x-cache header.
Can I bring my own provider keys?
Yes, free on every paid account. Route through your Anthropic / OpenAI / Google / etc. key and pay zero per-request fees on that traffic — your existing volume discounts apply.
Sub-accounts and per-user keys?
Yes — one API call mints a scoped key with its own spend cap, rate limit, default tag, and analytics. Ideal for marketplaces or multi-tenant apps that want to meter end users.
How do STT, TTS, embeddings, video, image price?
Every modality uses the same math: underlying model rate plus 5%. Whisper, ElevenLabs, BGE, Flux, Veo — all one simple line item.
Do credits expire?
Topup credits never expire — once you pay, the balance stays. The $1 card credit is the exception: it expires 30 days after you add a card. Refunds on unused topup balances within 90 days, no questions.
What's in Enterprise?
Evals, guardrails, replay + shadow A/B, and versioned prompt IDs — the gateway primitives. Plus SSO, 99.95% SLA, DPA, SOC 2, private audit export, dedicated endpoint, and a named engineer on Slack Connect. Details on /enterprise.

* Add a card (no charge) for $1 in credit on the selected models above. The $1 expires 30 days after you add the card. Topups never expire.