openai-compatible · 200+ models · wholesale rates

every model.
one key.
flat rate.

Route OpenAI, Anthropic, Gemini, DeepSeek, Kimi, Qwen and 200+ more through one OpenAI-compatible endpoint. We buy tokens at wholesale and pass through the price. Five-nines uptime backed by automatic provider failover.

Get an API key → See pricing

40%

avg savings vs official

99.95%

30-day uptime

200+

models routed

380ms

median ttft

drop-in compatible

Change one URL. Your code already works.

flatkey speaks the OpenAI Chat Completions API. Use any OpenAI SDK — just point its base_url at us.

curl

python

node

# works with the official openai SDK by changing base_url
curl https://api.flatkey.ai/v1/chat/completions \
  -H "Authorization: Bearer $FLATKEY_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-4-6",
    "messages": [{"role":"user","content":"Hello"}]
  }'

from openai import OpenAI

client = OpenAI(
    api_key=os.environ["FLATKEY_KEY"],
    base_url="https://api.flatkey.ai/v1",
)

resp = client.chat.completions.create(
    model="claude-sonnet-4-6",
    messages=[{"role": "user", "content": "Hello"}],
)
print(resp.choices[0].message.content)

import OpenAI from "openai";

const ai = new OpenAI({
  apiKey: process.env.FLATKEY_KEY,
  baseURL: "https://api.flatkey.ai/v1",
});

const r = await ai.chat.completions.create({
  model: "claude-sonnet-4-6",
  messages: [{ role: "user", content: "Hello" }],
});

pricing

Wholesale rates. No surge. No surprise.

We negotiate volume contracts with every upstream and pass the price through. The same number every month.

model

provider direct

openrouter

flatkey

save

gpt-4oopenai · 128k

$5.00 / $15.00

$3.20 / $10.40

−36%

claude-sonnet-4-6anthropic · 200k

$3.00 / $15.00

$1.95 / $10.05

−33%

gemini-2.5-progoogle · 1M

$1.25 / $10.00

$0.85 / $7.00

−31%

deepseek-v3.2deepseek · 128k

$0.27 / $1.10

$0.18 / $0.74

−33%

kimi-k2moonshot · 200k

$0.55 / $2.20

$0.36 / $1.45

−34%

Prices per 1M tokens (input / output). Rates updated weekly. See full price list →

why flatkey

Built for the line in your finance dashboard.

Wholesale, not retail.

We negotiate bulk contracts with every upstream and pass through cost + a thin margin. No reseller markup. No surge windows. The price you sign up at is the price you pay next month.

Failover, not finger-crossed.

When OpenAI 503s or Anthropic rate-limits you, we route the same request through a backup provider automatically. Your app sees one endpoint. We handle the rest.

→

One key. Every model.

OpenAI-compatible Chat Completions, Embeddings, and Vision. Use the OpenAI SDK you already have — change base_url to https://api.flatkey.ai/v1 and you're live.

models

Every model worth using.

200+ models from 12+ providers. New models added within 24 hours of release.

openai

gpt-4o · o3

anthropic

claude-opus-4-7

anthropic

claude-sonnet-4-6

google

gemini-2.5-pro

google

gemini-2.5-flash

deepseek

deepseek-v3.2

deepseek

deepseek-r1

moonshot

kimi-k2

alibaba

qwen-3-max

Wholesale price means nothing if it's down.

Every request goes through a 4-region edge with health checks every 30 seconds. We re-route around upstream incidents before your app's retry kicks in.

99.95%

30-day uptime

380ms

median ttft

1.8s

p99 latency

redundant routes

app.flatkey.ai/health →

every model.one key.flat rate.