x-portkey-config

Question

In such cases, instead of passing configID, you can pass raw config in the x-portkey-config header. Example curl header: --header 'x-portkey-config: {"cache": {"mode": "semantic"}}'

Saif · Answer

So here's what I tried:curl https://api.portkey.ai/v1/chat/completions \ -H "Content-Type: application/json" \ -H "x-portkey-api-key: apikeyiek2" \ -H "x-portkey-virtual-key: open-ai-key-virtual" \ -H "x-portkey-config: {"cache": {"mode": "semantic"}}" \ -d '{ "model": "gpt-3.5-turbo", "messages": [ { "role": "system", "content": "You are a helpful assistant." }, { "role": "user", "content": "Hello!" } ] }'And I got the following response{"status":"failure","message":"Invalid config passed. You need to pass a valid json"}

Saif · Answer

If I remove the x-portkey-config, then API call successfully flies. Let me know if you find something wrong with the Curl request.

visarg · Answer

This is happening because of wrong quotes used. x-portkey-config is surrounded with double quotes and the value of x-portkey-config also has double quotes inside of it. This fixed version should work where all the header quotes have been changes to single quote

Plain Text

curl https://api.portkey.ai/v1/chat/completions \
  -H 'Content-Type: application/json' \
  -H 'x-portkey-api-key: apikeyiek2' \
  -H 'x-portkey-virtual-key: open-ai-key-virtual' \
  -H 'x-portkey-config: {"cache": {"mode": "semantic"}}' \
  -d '{
    "model": "gpt-3.5-turbo",
    "messages": [
      { "role": "system", "content": "You are a helpful assistant." },
      { "role": "user", "content": "Hello!" }
    ]
  }'

Saif · Answer

Thank you 🙂

visarg · Answer

You're welcome. Closing this thread as its resolved.

Saif · Answer

Sorry for re-opening this thread, but I tried similarly with loadbalance

Plain Text

curl --request POST \
  --url https://api.portkey.ai/v1/chat/completions \
  --header 'Content-Type: application/json' \
  --header 'x-portkey-api-key: IxXxxxxxxxxHZxkAxxxxxxxx=' \
  --header 'x-portkey-config: {"strategy":{"mode":"loadbalance"},"targets":[{"virtualKey":"openai-virtual-key","weight":0.7},{"virtualKey":"azure-virtual-key","weight":0.3}]}' \
  --header 'x-portkey-virtual-key: open-ai-key-04ba3e' \
  --data '{
    "model": "gpt-3.5-turbo",
    "messages": [
      { "role": "system", "content": "You are a helpful assistant." },
      { "role": "user", "content": "Hello!" }
    ]
  }'

I have tried removing specific x-portkey-virtual-key too.

Saif · Answer

I get following response:{ "status": "failure", "message": "Invalid config passed", "errors": [ "path: targets,0, message: Invalid configuration. It must have either 'provider' and 'api_key', or 'strategy' and 'targets', or 'cache', or 'target'", "path: targets,1, message: Invalid configuration. It must have either 'provider' and 'api_key', or 'strategy' and 'targets', or 'cache', or 'target'" ]
}From this I understand this much - I can't use cache and loadbalance in the same config. Is my understanding correct?

Saif · Answer

I got this working 🟢 curl --request POST \ --url https://api.portkey.ai/v1/chat/completions \ --header 'Content-Type: application/json' \ --header 'x-portkey-api-key: apikey=' \ --header 'x-portkey-config: {"retry":{"attempts":3},"cache":{"mode":"simple"},"strategy":{"mode":"loadbalance"},"targets":[{"virtual_key":"open-ai-key","weight":0.7},{"virtual_key":"test-virtual-ke","weight":0.3}]}' \ --header 'x-portkey-virtual-key: open-ai-key' \ --data '{ "model": "gpt-3.5-turbo", "messages": [ { "role": "system", "content": "You are a helpful assistant." }, { "role": "user", "content": "Hello!" } ] }'

Welcome to Portkey Forum

x-portkey-config