Proxy service that add a Chain of Thought logic for any openAI compatible endpoints (like Groq <3)
curl http://localhost:3000/chat/completions -s \
-H "Content-Type: application/json" \
-H "Authorization: Bearer gsk_..." \
-d '{
"model": "llama3-8b-8192",
"messages": [{
"role": "user",
"content": "Explain the importance of fast language models"
}]
}'
Inspired by g1