Access GPT-5, Claude 4.7, Gemini, Grok 4, Mistral, DeepSeek and more through a single API. OpenAI-compatible. Pay per token with CYT.
curl https://api.chatyfai.com/v1/chat/completions \
-H "Authorization: Bearer sk-cyf-live-xxx" \
-H "Content-Type: application/json" \
-d '{
"model": "gpt-5",
"messages": [{"role": "user", "content": "Hello!"}]
}'Access OpenAI, Anthropic, Google, Mistral, DeepSeek and more through a single unified API.
Drop-in replacement for OpenAI SDK. Switch from OpenAI to ChatyfAI in one line of code.
No subscriptions. Pay only for what you use with ChatyfAI Tokens (CYT). Top up anytime.
Automatic key rotation, failover, and load balancing across provider API keys.
API key auth, IP whitelisting, rate limiting, budget caps. SOC2 ready.
Track usage, costs, and performance per model, per key. Webhooks for budget alerts.
1 CYT = 0.0001 EUR. Pay only for what you use.
| Model | Provider | Input (CYT/1K) | Output (CYT/1K) |
|---|---|---|---|
| GPT-5 | OpenAI | 25 | 100 |
| GPT-5 mini | OpenAI | 2 | 6 |
| Claude 4.7 Sonnet | Anthropic | 30 | 150 |
| Claude 4 Haiku | Anthropic | 3 | 12 |
| Gemini 2.5 Pro | 1 | 4 | |
| Grok 4 | xAI | 20 | 80 |
| Mistral Large | Mistral | 20 | 60 |
| DeepSeek V3 | DeepSeek | 1 | 3 |
ChatyfAI speaks the OpenAI Chat Completions protocol. Point any tool that accepts a custom base URL at https://api.chatyfai.com/v1 and you instantly get GPT-5, Claude 4.7, Gemini, Grok 4, Mistral, DeepSeek and Llama — billed per token in CYT, no per-tool subscription.
AI-first IDE. Add ChatyfAI as a custom OpenAI provider in Settings → Models.
// Cursor → Settings → Models → "Add custom OpenAI key"
// API Key sk-cyf-live-xxxxxxxxxxxxxxxxxxxxxxxx
// Base URL https://api.chatyfai.com/v1
// Model name gpt-5 (or claude-4-7-sonnet, gemini-2.5-pro, …)
// Then enable "Override OpenAI base URL".Open-source IDE assistant for VS Code & JetBrains. Add a single provider entry.
// ~/.continue/config.json
{
"models": [
{
"title": "ChatyfAI · GPT-5",
"provider": "openai",
"model": "gpt-5",
"apiKey": "sk-cyf-live-xxxxxxxxxxxxxxxxxxxxxxxx",
"apiBase": "https://api.chatyfai.com/v1"
},
{
"title": "ChatyfAI · Claude 4.7",
"provider": "openai",
"model": "claude-4-7-sonnet",
"apiKey": "sk-cyf-live-xxxxxxxxxxxxxxxxxxxxxxxx",
"apiBase": "https://api.chatyfai.com/v1"
}
]
}Autonomous coding agent for VS Code. Pick "OpenAI Compatible" as the provider.
// Cline → API Provider → "OpenAI Compatible"
// Base URL https://api.chatyfai.com/v1
// API Key sk-cyf-live-xxxxxxxxxxxxxxxxxxxxxxxx
// Model ID gpt-5
// Roo Code uses the same dialog; both honour streaming and tool calls.Terminal-native pair programmer. One CLI flag and you are routed through ChatyfAI.
# Aider — point any OpenAI-named model at ChatyfAI:
export OPENAI_API_BASE=https://api.chatyfai.com/v1
export OPENAI_API_KEY=sk-cyf-live-xxxxxxxxxxxxxxxxxxxxxxxx
aider --model gpt-5
aider --model claude-4-7-sonnet
aider --model deepseek-chat # cheap, fastTerminal coding agent. Drop the same OpenAI-compatible config into its providers file.
# opencode.json
{
"$schema": "https://opencode.ai/config.json",
"provider": {
"chatyfai": {
"options": {
"baseURL": "https://api.chatyfai.com/v1",
"apiKey": "sk-cyf-live-xxxxxxxxxxxxxxxxxxxxxxxx"
},
"models": {
"gpt-5": { "name": "ChatyfAI · GPT-5" },
"claude-4-7-sonnet": { "name": "ChatyfAI · Claude 4.7" }
}
}
}
}Already using openai-python or openai-node? Two lines of config — same calls.
# Python
from openai import OpenAI
client = OpenAI(
base_url="https://api.chatyfai.com/v1",
api_key="sk-cyf-live-xxxxxxxxxxxxxxxxxxxxxxxx",
)
resp = client.chat.completions.create(
model="gpt-5",
messages=[{"role": "user", "content": "Hello"}],
)Compatible with anything that speaks the OpenAI SDK. Got a tool that isn't listed? Set its OpenAI base URL to https://api.chatyfai.com/v1 and your ChatyfAI API key — that's the whole config.
Free tier available. No credit card required to start.
Create a Free ChatyfAI AccountThe unified AI platform — chat, deep search, image & video generation.
Deploy on-premise with Docker, mTLS, LDAP/SSO and full data sovereignty.
Free, Pro and Enterprise plans — pay-per-token with ChatyfAI Tokens (CYT).
Side-by-side multi-prompting across every supported AI model.
Need higher limits or dedicated support? Contact the ChatyfAI API team.