2,620 questions
- Bountied 0
- Unanswered
- Frequent
- Score
- Trending
- Week
- Month
- Unanswered (my tags)
0
votes
1
answer
20
views
JsonOutputParser in LangChain raises OutputParserException when model includes conversational text
I am using LangChain to extract structured data from a user query using gpt-3.5-turbo. I have defined a Pydantic object and am passing the format instructions to the prompt.
However, occasionally the ...
0
votes
0
answers
28
views
OpenAI image input token caching
OpenAI GPT-Image API: Image input tokens not caching despite documentation suggesting they should
Problem Summary
I'm using OpenAI's GPT-Image models (gpt-image-1.5, gpt-image-1, gpt-image-1-mini) to ...
0
votes
0
answers
21
views
passing fresh JWT to AzureOpenAI
I am attempting to pass a JWT to an AzureOpenAI call but the service responds with a 401. When I instantiate AzureOpenAI with a JWT-injected default header, it works. However, the JWT can become stale ...
Best practices
0
votes
2
replies
56
views
Is it possible to create a custom GPT and connect it to external services like Notion?
Is it possible to create a custom GPT and connect it with available connectors?
I want to create a GPT that can create and read documents in Notion, basically acting like a second memory/brain. The ...
-2
votes
0
answers
56
views
Anthropic model in open ai client [closed]
I have anthropic models in Google cloud project. How can we use the anthropic models with open ai client like we can with other gemini models
Tooling
0
votes
3
replies
93
views
Why is OpenAI’s ChatGPT SDK preferred at the enterprise level over Google Gemini?
I’ve noticed that many enterprises seem to adopt OpenAI’s ChatGPT SDK more frequently than Google Gemini or other AI platforms. From my understanding, both platforms provide similar foundational AI ...
1
vote
1
answer
401
views
Turn off gpt-5-nano reasoning
I am performing name/gender inference and I want to use the gpt-5-nano model because it is fast. The problem is that I can't seem to turn off reasoning even with the reasoning=None flag. I can't even ...
0
votes
0
answers
99
views
Open-WebUI Not Detecting Model from lama.cpp
I have Open-WebUI running in Docker and connected it to the lama.cpp server API.
I followed the instructions from this URL:
https://docs.openwebui.com/tutorials/integrations/deepseekr1-dynamic/#step-1-...
Advice
1
vote
2
replies
71
views
Unable to download package
I have issues with downloading "@stream-io/openai-realtime-api" package in NextJS project.
Error:
enter image description here
I tried to clear cash, change registry, used "--legacy-...
0
votes
0
answers
39
views
OpenAPI Spec client fail to compile with ambiguous types
I downloaded the OpenAI's OpenAPI Spec file and tried to generate and compile the client and faced a few issues and wonder if anyone knows the fix for it.
This is the specification: https://app....
Advice
2
votes
2
replies
78
views
RAG with Pinecone + GPT-5 for generating new math problems: incoherent outputs, mixed chunks, and lack of originality
I’m building a tool that generates new mathematics exam problems using an internal database of past problems.
My current setup uses a RAG pipeline, Pinecone as the vector database, and GPT-5 as the ...
0
votes
0
answers
45
views
Azure OpenAI model token rate limit header (x-ratelimit-limit-tokens) is -1
I am trying to get my current rate limit for an OpenAI model hosted on Azure. The response headers contain keys x-ratelimit-limit-tokens and x-ratelimit-remaining-tokens which are supposed to contain ...
3
votes
0
answers
97
views
Weird OpenAI streaming bug: concurrent JS streams randomly mix chunks across responses
I’m hitting a super weird issue with OpenAI’s streaming API.
I’m building a multi-message streaming thing in plain JS (no frameworks) where I need to pipe multiple OpenAI responses one after another. ...
-1
votes
1
answer
124
views
Open's AI MCP server redirect URI
I am configuring Open AI to talk to my MCP server , which uses SSO.
On the Chat GPT web UI, when adding a connector, by clicking on the connectors section and adding a new one:
I add the Client ID ...
Advice
0
votes
0
replies
82
views
Does OpenAI API TPM limit count input tokens, output tokens, or both?
I’m a bit confused about how OpenAI’s API rate limits work - specifically the TPM (tokens per minute) limit.
If I have, for example, 2 million TPM, is that limit calculated based on:
only the input ...