The AI Infrastructure
for Serious Builders.
One Unified API. Access Llama 3, Mistral, and GPT-4 without the headache.Smart Routing. 99.9% Uptime. Local Pricing.
❌ The Old Way
import openai
import anthropic
import cohere
# Managing 5 different keys
# Handling 5 different rate limits
# Dealing with sanctions/VPNs
if provider == 'openai':
client = OpenAI(api_key=...)
elif provider == 'anthropic':
client = Anthropic(...)
✅ The Thinx Way
import
openai
# One Client. Any Model.
client = OpenAI(
base_url="https://api.thethinx.ir/v1",
api_key="thinx-sk-..."
)
# Switch models instantly
response = client.chat.completions.create(
model="llama-3-70b-instruct",
messages=[...]
)
Smart Routing
We automatically route your request to the fastest and cheapest available GPU provider globally.
Unified Billing
Stop managing 10 credit cards. One invoice, transparent usage tracking per token.
Optimized for Iran
Low latency edge nodes. Payment in Rials. No VPN required for your servers.
Try it in your terminal
-H "Authorization: Bearer thinx-public-demo"