How It Works
FuturMix implements the OpenAI API specification — the same /v1/chat/completions endpoint format used by OpenAI. This means any code, SDK, or tool built for OpenAI works with FuturMix out of the box.
The difference: through FuturMix, you can access models from Anthropic, Google, DeepSeek, ByteDance, and Kimi — not just OpenAI — all through the same API key and endpoint.
⚡ One change, all models
Set base_url = "https://futurmix.ai/v1" and use model IDs like claude-sonnet-4-6, gpt-5.4, gemini-2.5-pro, or deepseek-chat-v3. Your existing OpenAI code works without any other changes.
API Endpoints
FuturMix supports three API formats. Use whichever your tools require:
OpenAI-Compatible
https://futurmix.ai/v1Works with: All OpenAI SDKs, Cursor, Aider, Continue, Cline, LangChain
Endpoints: /chat/completions, /responses, /models
Native Anthropic
https://futurmix.ai/anthropicWorks with: Claude Code, Anthropic SDK
Endpoints: /v1/messages, /v1/models
Native Gemini
https://futurmix.ai/geminiWorks with: Google AI SDK
Endpoints: /v1beta/models/:generateContent
Available Models
All models are accessible through the OpenAI-compatible endpoint. Use these model IDs in your API calls:
| Model ID | Provider | Input/1M | Output/1M | Context |
|---|---|---|---|---|
| claude-sonnet-4-6 | Anthropic | $2.70 | $13.50 | 200K |
| claude-opus-4-6 | Anthropic | $4.50 | $22.50 | 200K |
| claude-haiku-4-5-20251001 | Anthropic | $0.90 | $4.50 | 200K |
| gpt-5.4 | OpenAI | $1.75 | $10.50 | 200K |
| gpt-5.4-mini | OpenAI | $0.53 | $3.15 | 200K |
| gpt-5.4-nano | OpenAI | $0.14 | $0.88 | 200K |
| gemini-2.5-pro | $1.00 | $8.00 | 1M | |
| gemini-2.5-flash | $0.24 | $2.00 | 1M | |
| deepseek-chat-v3 | DeepSeek | $0.19 | $0.77 | 128K |
| deepseek-r1 | DeepSeek | $0.39 | $1.53 | 128K |
Prices include FuturMix discount (10-30% off official rates). View all 22+ models →
5-Minute Setup
Create your account
Sign up at futurmix.ai/register. No credit card required to start. Get your API key from the dashboard.
Set your base URL and API key
Replace OpenAI's URL with FuturMix's. Everything else stays the same.
# Environment variables (works with most tools)
export OPENAI_API_BASE="https://futurmix.ai/v1"
export OPENAI_API_KEY="sk-your-futurmix-key"
Choose a model and make your first call
Use any model ID from the table above. Here's a complete example:
Python
from openai import OpenAI
client = OpenAI(
base_url="https://futurmix.ai/v1",
api_key="sk-your-futurmix-key",
)
# Use Claude through OpenAI SDK
response = client.chat.completions.create(
model="claude-sonnet-4-6",
messages=[
{"role": "system", "content": "You are a helpful assistant."},
{"role": "user", "content": "Explain quantum computing in 3 sentences."},
],
max_tokens=500,
)
print(response.choices[0].message.content)
Node.js / TypeScript
import OpenAI from 'openai';
const client = new OpenAI({
baseURL: 'https://futurmix.ai/v1',
apiKey: 'sk-your-futurmix-key',
});
const response = await client.chat.completions.create({
model: 'gemini-2.5-pro',
messages: [{ role: 'user', content: 'Hello' }],
});
cURL
curl https://futurmix.ai/v1/chat/completions \
-H "Authorization: Bearer sk-your-futurmix-key" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek-chat-v3",
"messages": [{"role": "user", "content": "Hello"}],
"stream": true
}'
Streaming (Python)
# Streaming works exactly like OpenAI
stream = client.chat.completions.create(
model="claude-sonnet-4-6",
messages=[{"role": "user", "content": "Write a short poem"}],
stream=True,
)
for chunk in stream:
if chunk.choices[0].delta.content:
print(chunk.choices[0].delta.content, end="")
Tool Integration Guide
Here's how to configure popular developer tools to use FuturMix:
Claude Code
# Uses native Anthropic format
export ANTHROPIC_BASE_URL="https://futurmix.ai/anthropic"
export ANTHROPIC_API_KEY="sk-your-futurmix-key"
Cursor
# Settings > Models > OpenAI API Key
# Base URL: https://futurmix.ai/v1
# API Key: sk-your-futurmix-key
# Model: claude-sonnet-4-6 (or any model ID)
Aider
export OPENAI_API_BASE="https://futurmix.ai/v1"
export OPENAI_API_KEY="sk-your-futurmix-key"
aider --model claude-sonnet-4-6
Continue (VS Code)
// .continue/config.json
{
"models": [{
"provider": "openai",
"model": "claude-sonnet-4-6",
"apiBase": "https://futurmix.ai/v1",
"apiKey": "sk-your-futurmix-key"
}]
}
LangChain
from langchain_openai import ChatOpenAI
llm = ChatOpenAI(
base_url="https://futurmix.ai/v1",
api_key="sk-your-futurmix-key",
model="claude-sonnet-4-6",
)
Error Handling
FuturMix returns standard OpenAI-format error responses. Your existing error handling code works without changes.
| HTTP Code | Meaning | Action |
|---|---|---|
400 | Bad request (invalid params) | Check request body format |
401 | Invalid API key | Check your API key in the dashboard |
402 | Insufficient credits | Top up your account balance |
429 | Rate limited | Back off and retry (auto-failover may handle this) |
500 | Server error | Retry with exponential backoff |
503 | Model temporarily unavailable | Try a different model or wait |
# Error handling example
try:
response = client.chat.completions.create(
model="claude-sonnet-4-6",
messages=[{"role": "user", "content": "Hello"}],
)
except openai.AuthenticationError:
print("Invalid API key")
except openai.RateLimitError:
print("Rate limited — retrying...")
except openai.APIError as e:
print(f"API error: {e}")
Frequently Asked Questions
What is an OpenAI-compatible API?
An OpenAI-compatible API implements the same endpoint format as OpenAI (
/v1/chat/completions) but lets you access models from multiple providers. You use the same OpenAI SDK — just change the base_url to point to the alternative provider.Can I use Claude through an OpenAI-compatible API?
Yes. FuturMix provides all Claude models (Sonnet 4, Opus 4.6, Haiku 4.5) through its OpenAI-compatible endpoint at
https://futurmix.ai/v1. Use the OpenAI SDK with model names like claude-sonnet-4-6. FuturMix also supports the native Anthropic API at /anthropic for tools like Claude Code.Does FuturMix support the native Anthropic API format too?
Yes. FuturMix supports three API formats: OpenAI-compatible (
/v1), native Anthropic Messages (/anthropic/v1/messages), and native Gemini GenerateContent (/gemini/v1beta/models). Use whichever format your tools require — same API key works for all three.What tools work with FuturMix's OpenAI-compatible API?
Any tool that supports OpenAI's API format: Claude Code, Cursor, Aider, Cline, Continue, Windsurf, LangChain, LlamaIndex, and any custom code using the OpenAI SDK for Python, Node.js, Go, or PHP.
How do I handle errors?
FuturMix returns standard OpenAI-format error responses with HTTP status codes (400, 401, 429, 500) and error objects. Your existing OpenAI error handling code works without changes. FuturMix also provides automatic failover — if one provider endpoint fails, requests are routed to a backup.
Is there a rate limit?
FuturMix does not enforce hard rate limits for paid accounts. If you hit provider-level limits, FuturMix's auto-failover routes to an alternative endpoint. For enterprise volume requirements, contact admin@futurmix.ai.
Start Using 22+ AI Models Through One API
Get a free API key, set your base URL, and access Claude, GPT, Gemini, and DeepSeek in under 5 minutes.