On Azure AI foundry, Prompt flow is unusable with O1 models because it doesn't have the 'max_completion_tokens' parameter on the UI

Question

We use Azure AI Foundry's PromptFlow for internal RAG and so far the O1 and O3 models have not worked simply because parameter name was changed and not added to Promptflow. Kindly suggest a workaround.

Are there examples you can provide for how to use o1 and o3-mini in promptflow UI on Azure Ai foundry ?

Answer 1

Hi Subhrajit Bhowmik!

I'm glad that you were able to resolve your issue and thank you for posting your solution so that others experiencing the same thing can easily reference this! Since the Microsoft Q&A community has a policy that "The question author cannot accept their own answer. They can only accept answers by others ", I'll repost your solution in case you'd like to accept the answer.

Ask: Not able to use O1 or O3 mini in prompt tools because it only takes max_completion_tokens instead of supported param max_tokens.

Solution: The issue was resolved by creating a custom python tool using OpenAI and langchain as shown in below screenshot.

If I missed anything please let me know and I'd be happy to add it to this answer, or feel free to comment below with any additional information.

If you have any other questions, please let me know. Thank you again for your time and patience throughout this issue.

Please don’t forget to Accept Answer and Yes for "was this answer helpful" wherever the information provided helps you, this can be beneficial to other community members.

Thank You.