Proxy Mode Issues with Together.ai as the Backend Service #2027
Unanswered
vitalii-lobanov
asked this question in
Q&A
Replies: 2 comments
-
Hi @vitalii-lobanov your settings look good to me, can we hop on a quick call and debug this together ? Sharing a link to my cal for your convenience: |
Beta Was this translation helpful? Give feedback.
0 replies
-
I've identified and corrected an error in my configuration. Initially, I had set the API base URL for Together.ai to https://api.together.xyz/inference, which was incorrect. After updating it to https://api.together.xyz/v1, everything is now functioning properly. Here is my updated configuration:
|
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
I'm encountering an error when attempting to use Litellm in proxy mode, with Together.ai serving as the backend service. Here is the curl query:
The response is the following:
Litellm's logs:
Below is the complete Litellm test configuration that I utilized for this experiment:
Litellm version:
I am able to access Together.ai's API from the same machine where Litellm is running. Here is an example of the query:
The response is:
Additionally, I've verified Litellm's functionality by testing it with the Openrouter API, which worked correctly. Here is the query used for testing:
The response is correct:
Finally, I will provide the complete Litellm logs corresponding to the error encountered when Litellm is used as a proxy for Together.ai's API:
The question arises as to what might have been configured incorrectly, and how can I correctly set up Litellm to serve as a proxy for Together.ai's service?
Beta Was this translation helpful? Give feedback.
All reactions