25 models · OpenAI-compatible API

Every model.One key.

Claude Opus 5.5, GPT-6 Astra, Gemini, Grok, DeepSeek, Kimi and other top models behind one API — cheaper than official prices. Change the base_url in your code; everything else already works.

Get an API keyBrowse models
base_urlhttps://api.flua.ink/v1
models
25

models

providers
11

providers

biggest saving vs official prices
−80%

biggest saving vs official prices

import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.flua.ink/v1",
    api_key=os.environ["FLUA_API_KEY"],
)

stream = client.chat.completions.create(
    model="claude-opus-5-5",
    messages=[
        {"role": "user", "content": "Write a haiku about flight"},
    ],
    stream=True,
)
Response· claude-opus-5-5

200 OK16 → 24 tokens$0.00022

Models from the leading labs, in one catalog

  • Anthropic
  • OpenAI
  • Google
  • xAI
  • DeepSeek
  • Alibaba
  • Moonshot AI
  • Zhipu AI
  • MiniMax
  • Xiaomi
  • Tencent

Three steps to your first response

  1. 01

    Create a key

    Sign up and issue an API key in your dashboard. Make as many as you need — one per project.

  2. 02

    Top up your balance

    One balance for every model. No subscription: you only pay for the tokens you actually use.

  3. 03

    Swap the base_url

    The official OpenAI and Anthropic SDKs work as is — just point them at flua with your key.

app.py−2 +2
from openai import OpenAI
 
client = OpenAI(
-    base_url="https://api.openai.com/v1",
+ 
-    api_key="sk-proj-••••••••",
+ 
)

Everything you need to build with models, in one API

  • POST/v1/chat/completionsOpenAI
  • POST/v1/responsesOpenAI
  • POST/v1/messagesAnthropic
  • POST/v1/images/generationsOpenAI
  • GET/v1/modelsOpenAI

OpenAI & Anthropic compatible

Chat Completions, Responses and the Messages API. Connect from any language or framework with the code you already have.

Balance

$48.204 049 ₽2 169 ₴

One balance

Top up once and spend it on any model. Every model's price is published upfront.

data: {"delta":{"content":"A"}}

data: {"delta":{"content":" swift"}}

data: {"delta":{"content":" over"}}

data: {"delta":{"content":" water"}}

data: {"delta":{"content":" —"}}

data: {"delta":{"content":"\n"}}

data: {"delta":{"content":"one"}}

Streaming

Tokens arrive as they are generated over Server-Sent Events — just like the official APIs.

gpt-image-2.5-flare
nano-banana-pro
grok-imagine-image
nano-banana-2

Images

GPT Image, Nano Banana and Grok Imagine — image generation with the same key.

  • production

    sk-flua-…9f2a

    $200.0016 800 ₽9 000 ₴ limit

  • staging

    sk-flua-…41c7

    $25.002 100 ₽1 125 ₴ limit

  • laptop

    sk-flua-…e0b3

    no limit

Keys under control

A separate key per project, a spending limit per key and one-click revoke.

14:02:18claude-opus-5-51,204 → 386200$0.00310,26 ₽0,14 ₴
14:02:11gpt-6-astra842 → 512200$0.00270,23 ₽0,12 ₴
14:01:57gemini-3.8-flash3,310 → 140200$0.00690,58 ₽0,31 ₴
14:01:40deepseek-v4.1-flash620 → 1,024200$0.00330,28 ₽0,15 ₴

Transparent usage

Every request is logged with model, tokens and cost. See exactly where the money goes.

The flagship of every lab

The best models from Anthropic, OpenAI, Google, xAI and more — no separate accounts or subscriptions.

All 25 models

* per 1M tokens

Works with the tools you already use

Any client that lets you set a custom OpenAI- or Anthropic-compatible endpoint.

Setup guides

Pay only for tokens

No plans, no subscriptions. Top up your balance and it is spent at each model's price.

  • No monthly fee
  • Per-model price per 1M tokens
  • Pay in dollars, rubles or hryvnias
  • Charged for actual usage only
See all prices

Cost calculator

1,500
500
200
One request
$0.007
Per day
$1.40
Per month
$42.00

Questions & answers

What is flua?

An API gateway to models from different labs: Anthropic, OpenAI, Google, xAI, DeepSeek and others. You get one key and one balance, and pick the model with the model parameter of each request.

Do I need to rewrite my code?

No. flua is compatible with the OpenAI API (Chat Completions and Responses) and with the Anthropic Messages API. Swap the base_url and the key — the official SDKs keep working.

How is usage priced?

Per token: input tokens (your request) and output tokens (the model's answer) are priced separately, per 1M tokens. Image generation is priced per request. The exact token counts come back in the usage field of every response.

How do I top up?

In your dashboard (Balance) or in our Telegram bot: with crypto through CryptoBot — USDT, TON, BTC and more — or with a top-up code. The money lands on your balance automatically.

Can I cap my spending?

Yes. Each API key can have a spending limit: once it is reached, that key stops accepting requests while the others keep working.

Are streaming and tool calling supported?

Yes. Streaming over Server-Sent Events and tool calling work exactly as in the official APIs — for the models that support them.

Which models are available?

The catalog currently lists 25 models from 11 providers. The up-to-date list is always on the models page and at GET /v1/models.

Start in a minute

Sign up, grab a key, and send your first request to any model.

Get an API keyQuickstart