How much does Requesty cost?
Requesty charges each model's provider list price plus a 5% fee on every top-up, so a $100 top-up costs $105, and its free plan covers free models only at 200 requests a day. A coding agent on Claude Sonnet 5.5 costs about $74 a month through Requesty on typical working days and $608 when every day is busy. Standard Compute is the cost-optimized alternative for coding agents: a flat plan from $19 with included compute, and a router that sends routine steps to efficient models.
What you pay on Requesty
| Item | Price | Detail |
|---|---|---|
| Tokens | Provider list price | Claude Sonnet 5.5 is $2 input and $10 output per million tokens, the same as Anthropic's price |
| Gateway fee | 5% of every top-up | Charged when you add credit: "a $100 top-up costs $105 and credits $100" |
| Free plan | $0 | Free models only, 200 requests a day, no card |
| Your own provider keys | 0% for startups after six months | Supported for OpenAI, Anthropic, Google, Mistral and xAI. The pricing page lists BYOK without a fee figure |
| EU endpoint | Included on every plan | Frankfurt. EU-hosted model versions cost more: Sonnet 5.5 @eu is $2.20 and $11 |
| Enterprise | Custom | Group budgets, custom routing rules, invoicing |
From requesty.ai/pricing, requesty.ai/startup and docs.requesty.ai, checked October 13, 2026. Startups backed by YC, 20VC, Tapestry or TinyVC can get $1,000 in credits and six months with no fee.
What a coding-agent month costs
Pinned to one Claude model, a typical month costs less on Requesty and a busy month less on a Standard Compute plan, which covers about 19 of its 20 days. Through any gateway that resells at list price, the lever on the bill is which model handles each request. A hundred typical coding-agent requests cost $5.33 on Claude Sonnet 5.5 and $0.41 on GPT-6 Luna at list price. Requesty leaves that choice to you; Standard Compute makes it per request.
| Month of coding-agent use | Requesty (list + 5%) | Standard Compute plan |
|---|---|---|
| Typical days, Claude Sonnet 5.5 | $74 | Standard, $89 with a $95 budget |
| Typical days, Claude Opus 5.5 | $148 | Pro, $249 with a $269 budget |
| Busy days, Claude Sonnet 5.5 | $608 | Pro Plus, $499 with a $549 budget (about 19 busy days) |
| Busy days, Claude Opus 5.5 | $1,217 | Growth, $999 with a $1,119 budget (about 19 busy days) |
| Typical days, GPT-6 Luna | $5 | Starter, $19 with a $20 budget |
Month = 20 working days. Typical is the median active day in our Agent Burn Index, busy the 90th percentile, from coding agents such as OpenCode, Kilo and Roo on our gateway, September 13 to October 12, 2026; Claude Code and Codex were not in the sample. Standard Compute budgets assume every request on that one model; routing routine steps to efficient models makes a budget last longer.
Routing: who picks the model
Requesty's routing policies are fallback, load balancing and latency, plus managed policies that fail over between providers of one model. Its docs answer "Does latency routing consider cost?" with "No", and suggest load balancing toward cheaper models or ordering a fallback chain by price. That works when you know which requests are easy. A coding agent mixes file reads, edits and planning in one session, which is why Standard Compute classifies each request and sends the routine ones to efficient models.
Spend limits and what happens at zero
- Each Requesty API key can have a monthly spend cap; when it is reached, "new requests for that key are blocked until the next billing period".
- Monthly limits per user come with every plan. Group budgets need Enterprise.
- When the prepaid balance runs out, requests are rejected with HTTP 402 until you top up, or auto top-up refills it.
- Requesty does not cap requests per minute. It caps requests in flight at once, scaled to your balance: five times the balance in dollars below $250 of lifetime spend.
- On Standard Compute the plan's budget is the cap: once a key's spend reaches it, the next request gets HTTP 402 until renewal, an upgrade or a top-up you choose, with no overage.
The catch
Requesty's 5% is small, and the bigger number is the token bill underneath it. Every step of an agent resends its context at list price, so a busy month on Claude Opus 5.5 passes $1,200 whichever gateway you use, unless routine steps move to cheaper models.
Where Standard Compute fits
Standard Compute is the cost-optimized option for coding agents: a flat plan from $19 a month with included compute that stops at its budget, smart routing that sends routine steps to efficient models, a plan per key for teams and agencies, Claude Code on /v1/messages and Codex on /v1/responses, and Europe or US rules on any plan.
FAQ
Does Requesty charge a markup?
Yes, 5%. Token prices match the provider's list price, and Requesty adds 5% to each top-up: a $100 top-up costs $105 and credits $100.
Is Requesty free?
The free plan is free models only, with a limit of 200 requests a day and no credit card. Paid models need pay-as-you-go credit.
Is Requesty cheaper than OpenRouter?
Slightly, per dollar of tokens: Requesty adds 5% to top-ups and OpenRouter charges a 5.5% fee on card credit purchases, with a $0.80 minimum. Both resell at provider list price, so the model you run matters far more than the fee.
What does Requesty cost for Claude Code?
Claude Sonnet 5.5 is $2 input and $10 output per million tokens, plus the 5% fee. On our measured coding-agent traffic that is about $5.60 per 100 requests, or $74 for a month of typical working days.
Does Requesty have an EU endpoint?
Yes, router.eu.requesty.ai in Frankfurt, on every plan including Free. Requesty's docs say you must also choose an EU-hosted model, and those versions cost about 10% more.