One Flatkey API
No provider credentials scattered across agents, repos, and environments.
One API. Every Models. Lower Price.
Built for developers, startups, and AI teams. Access 100+ official AI models and 1,000+ pay-per-call tools from one account — one balance, one invoice, one SLA. Failed calls cost you nothing.
Claude Opus 5.5Input $3.2 · Output $16 / 1M tokens
gpt-6-astraInput $8 · Output $40 / 1M tokensgpt-6-solInput $1.6 · Output $8 / 1M tokensgpt-6-lunaInput $0.08 · Output $0.4 / 1M tokens
typesafe/jev-1.13Input $0.0336 · Output $0 / 1M tokens
deepseek-v4-1-flashInput from $0.12 · Output from $0.48 / 1M tokensEvery model,100+ Every tool,1000+ One key.
Unified model + tool router
Use one Flatkey interface for leading AI models, social platforms, web data, crawlers, GTM intelligence, and more.
All ModelsNormalized calls and results
Consistent schemas, error envelopes, request IDs, and usage records.
Budgets and allowlists
Control which agents can call which tools and how much they may spend.
Pay for successful work
Models and tools share one balance; failed Flatkey-side calls cost nothing.
One key, every model, your choice.
Get one key
Sign up, grab a single Flatkey API, and replace your base URL. Your existing OpenAI SDK just works.
Choose your model
Set the model ID explicitly for each request. Switch providers without adding another account or integration.
Pay for what worked
Models and tools share one prepaid balance. If a call fails on our side, you don't pay for it. It's in the SLA.
Below-official pricing, plainly stated
Discounts come from the live model directory. Official is the 100% baseline; Flatkey shows the current billed share and savings for the four most-discounted featured models.
View All Model PricesWhy choose Flatkey?
Flatkey mesh 99.98%
Stability you can hold us to.
5Channel Classes, Automatic Failover
$0Balance Consumed By Flatkey-Side Errors
One request, five upstream channel classes. When a channel degrades, the router reroutes automatically — and the SLA is a public document, not a sales promise.
Successful request
req_8F2A$0.83Flatkey-side failure
req_8F2B$0.00One invoice
Models + toolsReadyEvery charge is traceable.
100%Request-level records
$0Flatkey-side failures
1Balance and invoice
Models and tools share one balance, while every successful request gets its own request ID and usage record. Flatkey-side failures are charged at $0.
ClaudeOpenAI
Kimi
SeedanceDeepseek
GoogleOfficial endpoints, verified hourly
Every request hits the real lab API — no fp8 re-serves, no silent swaps. 100+ models probed against official fingerprints, on a public log.
Sub-key capsSOC 2 Type IIModel allowlistsISO 27001Ledger APIGDPRInvoices 48hZero retention
Enterprise-grade governance
Tree-structured sub-keys with budgets and model allowlists, a per-request ledger API, invoices in 48h — on audited infrastructure.
1,300+ customers trust Flatkey with the hard stuff
Flatkey’s stability and high-quality support enable our team to focus more on product innovation, rather than on cumbersome operational overhead.
Flatkey’s unified model API enables our users to go live faster with production-grade latency and reliability.
Flatkey’s optimized inference lets creators seamlessly access the latest SOTA models in ComfyUI with zero infrastructure overhead.
Builder
Flatkey — lighter to deploy, faster to launch.
A single API that supports all access methods — call 100+ models directly from your code, agent, or terminal.
Call any model with an OpenAI-compatible API. Copy a ready-to-run example for your model and language.
View DetailsUse the OpenAI SDK you already know — with Flatkey as the gateway.
View DetailsGenerate images and videos from your terminal. Let your AI assistant drive the workflow.
View DetailsConnect your coding agent with one command, then use Flatkey from your existing projects.
View Detailscurl https://router.flatkey.ai/v1/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer sk-***" \
-d '{"model":"gpt-4o-mini","messages":[{"role":"user","content":"Say hello in one sentence."}]}'Questions? Talk to us.
Integration, model choice, or billing — online support and usage consultation for everything on flatkey.
Discord Community
discord.gg/VrbZFDXj5gBuilders answering builders — the team is in the channels daily
Reply within 1 business day
X (Twitter)
@flatkey101Follow or DM
Livechat
Start Chatting →Fastest answer, on the site
More AI, less cost on every official model
Whether you ship an agent today or route billions of tokens a month — one key, a signed SLA, and prices that drop as you grow.