# LLMs and AI

**URL:** https://discuss.streamlit.io/c/llm/26.md

[Latest](https://discuss.streamlit.io/latest.md) · [Categories](https://discuss.streamlit.io/categories.md) · [Tags](https://discuss.streamlit.io/tags.md)

---

## [About the LLMs and AI category](https://discuss.streamlit.io/t/about-the-llms-and-ai-category/42539)

<div class="topic-metadata">

**Author:** [@Caroline](https://discuss.streamlit.io/u/Caroline)\
**Replies:** 0

</div>

Questions about using large language models with Streamlit

---

## [Handling an AI Agent triggered MCP Elicitation callback](https://discuss.streamlit.io/t/handling-an-ai-agent-triggered-mcp-elicitation-callback/119502)

<div class="topic-metadata">

**Author:** [@Ben20](https://discuss.streamlit.io/u/Ben20)\
**Replies:** 2\
**Last updated:** [November 5, 2026, 5:54pm UTC](https://discuss.streamlit.io/t/handling-an-ai-agent-triggered-mcp-elicitation-callback/119502 "2026-11-05T17:54:03Z")

</div>

I’m wondering if anyone has an example they could share of handling an AI Agent triggered MCP Elicitation callback. I realize this might not be the best fit for Streamlit given its linear execution model, but I’m hoping …

---

## [Hello Streamlit Community! Exploring AI-Powered Data Apps](https://discuss.streamlit.io/t/hello-streamlit-community-exploring-ai-powered-data-apps/122947)

<div class="topic-metadata">

**Author:** [@Ankit35](https://discuss.streamlit.io/u/Ankit35)\
**Replies:** 1\
**Last updated:** [October 27, 2026, 4:20am UTC](https://discuss.streamlit.io/t/hello-streamlit-community-exploring-ai-powered-data-apps/122947 "2026-10-27T04:20:45Z")

</div>

Hi everyone! I’m Ankit Shekhawat, and I’m interested in AI development, Python-based applications, and interactive data experiences. I’m exploring how Streamlit can be used to build practical applications such as AI-po…

---

## [How to create a chat history on the side bar just like chatGPT?](https://discuss.streamlit.io/t/how-to-create-a-chat-history-on-the-side-bar-just-like-chatgpt/59492)

<div class="topic-metadata">

**Author:** [@lz-chen](https://discuss.streamlit.io/u/lz-chen)\
**Replies:** 20\
**Last updated:** [September 25, 2026, 6:46pm UTC](https://discuss.streamlit.io/t/how-to-create-a-chat-history-on-the-side-bar-just-like-chatgpt/59492 "2026-09-25T18:46:38Z")

</div>

Hi, I’m building a chat app with Streamlit and want to make it really chatGPT like. I haven’t found out an easy way to show the chat history (clickable and on click it will jump to that chat with possibility to continue …

---

## [Let's make a prompt library for Cursor & Streamlit](https://discuss.streamlit.io/t/lets-make-a-prompt-library-for-cursor-streamlit/79937)

<div class="topic-metadata">

**Author:** [@truevis](https://discuss.streamlit.io/u/truevis)\
**Replies:** 7\
**Last updated:** [July 23, 2026, 5:08am UTC](https://discuss.streamlit.io/t/lets-make-a-prompt-library-for-cursor-streamlit/79937 "2026-07-23T05:08:38Z")

</div>

Whereas Cursor + LLMs is really good at writing code, we can make .md files to reuse prompts to make the coding process faster. See this video: x.com Example \<make\_requirements.md\> : Create a \`requirements.txt\` file f…

---

## [Open source project Clinic AI Pro - How to integrate AI for free despite region block?](https://discuss.streamlit.io/t/open-source-project-clinic-ai-pro-how-to-integrate-ai-for-free-despite-region-block/121318)

<div class="topic-metadata">

**Author:** [@mouna](https://discuss.streamlit.io/u/mouna)\
**Replies:** 0\
**Last updated:** [April 20, 2026, 8:25am UTC](https://discuss.streamlit.io/t/open-source-project-clinic-ai-pro-how-to-integrate-ai-for-free-despite-region-block/121318 "2026-04-20T08:25:55Z")

</div>

Hello, I am the developer of an open-source project called “Clinic AI Pro” – an AI-powered clinic management platform supporting Arabic language. My problem: Claude API is not available in my region (error: “App unav…

---

## [Downloadable Streamlit API documentation for LLMs](https://discuss.streamlit.io/t/downloadable-streamlit-api-documentation-for-llms/120289)

<div class="topic-metadata">

**Author:** [@LeChuck](https://discuss.streamlit.io/u/LeChuck)\
**Replies:** 4\
**Last updated:** [January 9, 2026, 2:28am UTC](https://discuss.streamlit.io/t/downloadable-streamlit-api-documentation-for-llms/120289 "2026-01-09T02:28:32Z")

</div>

Hi Everyone, I am not sure if this is the right sub-forum, but I was wondering if there is downloadable version of the API documentation somewhere? Preferable as txt or pdf. I am asking since I am using LLMs like Claud…

---

## [.cursorrules for deprecated use\_container\_width](https://discuss.streamlit.io/t/cursorrules-for-deprecated-use-container-width/119576)

<div class="topic-metadata">

**Author:** [@truevis](https://discuss.streamlit.io/u/truevis)\
**Replies:** 0\
**Last updated:** [September 25, 2025, 6:45pm UTC](https://discuss.streamlit.io/t/cursorrules-for-deprecated-use-container-width/119576 "2025-09-25T18:45:22Z")

</div>

I came up with this rule to help LLMs write code without the warning: 2025-09-25 12:18:38.722 Please replace use\_container\_width with width. use\_container\_width will be removed after 2025-12-31. For use\_container\_width=…

---

## [fastai users](https://discuss.streamlit.io/t/fastai-users/118362)

<div class="topic-metadata">

**Author:** [@Asrorbek](https://discuss.streamlit.io/u/Asrorbek)\
**Replies:** 0\
**Last updated:** [August 6, 2025, 5:41pm UTC](https://discuss.streamlit.io/t/fastai-users/118362 "2025-08-06T17:41:04Z")

</div>

I have installed streamlit via pip and I can see it listed in \`pip list However, in VS Code, I get this error: (Import “streamlit” could not be resolved PylancereportMissingImports)

---

## [Why Isn’t My Streamlit App Handling Long LLM Conversations Smoothly?](https://discuss.streamlit.io/t/why-isn-t-my-streamlit-app-handling-long-llm-conversations-smoothly/117921)

<div class="topic-metadata">

**Author:** [@ricaxa](https://discuss.streamlit.io/u/ricaxa)\
**Replies:** 0\
**Last updated:** [July 28, 2025, 5:45am UTC](https://discuss.streamlit.io/t/why-isn-t-my-streamlit-app-handling-long-llm-conversations-smoothly/117921 "2025-07-28T05:45:00Z")

</div>

Hello I am building a chatbot-style app in Streamlit using the OpenAI API. It works well for short exchanges but once the conversation gets longer (6–8 turns); the app becomes laggy or occasionally freezes. :innocent: I…

---

## [Getting an error when deploying Streamlit app for LangChain-based RAG application](https://discuss.streamlit.io/t/getting-an-error-when-deploying-streamlit-app-for-langchain-based-rag-application/116458)

<div class="topic-metadata">

**Author:** [@prateek.m](https://discuss.streamlit.io/u/prateek.m)\
**Replies:** 1\
**Last updated:** [July 22, 2025, 9:35am UTC](https://discuss.streamlit.io/t/getting-an-error-when-deploying-streamlit-app-for-langchain-based-rag-application/116458 "2025-07-22T09:35:02Z")

</div>

I am trying to create an context-based retriever chat assistant using LangChain and RAG and deploying the interface via Streamlit. But I am getting the following error, after I upload the .docx file and enter the query: …

---

## [Unable to get streamlit page.](https://discuss.streamlit.io/t/unable-to-get-streamlit-page/116974)

<div class="topic-metadata">

**Author:** [@kish12](https://discuss.streamlit.io/u/kish12)\
**Replies:** 1\
**Last updated:** [July 11, 2025, 9:56pm UTC](https://discuss.streamlit.io/t/unable-to-get-streamlit-page/116974 "2025-07-11T21:56:32Z")

</div>

Hello, I am using Streamlit as a UI for my RAG agent. Streamlit APP URL accessable when we connected to a VPN and not working with different VPN. It has issue downloading index.B-cSXLfy.js. This file is partially downl…

---

## [What are the best practices for managing AI visibility in user-facing AI tools?](https://discuss.streamlit.io/t/what-are-the-best-practices-for-managing-ai-visibility-in-user-facing-ai-tools/115986)

<div class="topic-metadata">

**Author:** [@Olivia3](https://discuss.streamlit.io/u/Olivia3)\
**Replies:** 0\
**Last updated:** [June 23, 2025, 3:26pm UTC](https://discuss.streamlit.io/t/what-are-the-best-practices-for-managing-ai-visibility-in-user-facing-ai-tools/115986 "2025-06-23T15:26:07Z")

</div>

Hi everyone, I’m currently working on a Streamlit app powered by LLMs, and I’m exploring the best ways to make AI involvement clear to users. AI visibility is important for building trust, especially in tools where user…

---

## [Is LangGraph unable to access st.experimental\_user.email? How to provide this access](https://discuss.streamlit.io/t/is-langgraph-unable-to-access-st-experimental-user-email-how-to-provide-this-access/115739)

<div class="topic-metadata">

**Author:** [@WaqarQureshii](https://discuss.streamlit.io/u/WaqarQureshii)\
**Replies:** 3\
**Last updated:** [June 20, 2025, 11:02pm UTC](https://discuss.streamlit.io/t/is-langgraph-unable-to-access-st-experimental-user-email-how-to-provide-this-access/115739 "2025-06-20T23:02:40Z")

</div>

Github Link Error I am receiving: When trying to access the database via the chatbot in create\_resume, langgraph correctly calls my function setup as a tool: get\_user\_info, however it is unable to get the st.experimenta…

---

## [Streamlit chat app: each agent and user proxy's output is being printed 4 times in the web page](https://discuss.streamlit.io/t/streamlit-chat-app-each-agent-and-user-proxys-output-is-being-printed-4-times-in-the-web-page/55056)

<div class="topic-metadata">

**Author:** [@Bhavneesh\_Sharma](https://discuss.streamlit.io/u/Bhavneesh_Sharma)\
**Replies:** 5\
**Last updated:** [May 27, 2025, 9:51am UTC](https://discuss.streamlit.io/t/streamlit-chat-app-each-agent-and-user-proxys-output-is-being-printed-4-times-in-the-web-page/55056 "2025-05-27T09:51:50Z")

</div>

I am creating a chat app using Streamlit and Autogen library for our firm’s internal use by our employees. When I run the app using streamlit run app.py command after entering the agent and user proxy details in the web …

---

## [NotImplementedError: Cannot copy out of meta tensor; no data! when i am deploying my ML app on streamlit i am facing this issue](https://discuss.streamlit.io/t/notimplementederror-cannot-copy-out-of-meta-tensor-no-data-when-i-am-deploying-my-ml-app-on-streamlit-i-am-facing-this-issue/61709)

<div class="topic-metadata">

**Author:** [@chirag\_aswal](https://discuss.streamlit.io/u/chirag_aswal)\
**Replies:** 2\
**Last updated:** [May 14, 2025, 7:17am UTC](https://discuss.streamlit.io/t/notimplementederror-cannot-copy-out-of-meta-tensor-no-data-when-i-am-deploying-my-ml-app-on-streamlit-i-am-facing-this-issue/61709 "2025-05-14T07:17:35Z")

</div>

I am getting this error when i am deploying my LLM on streamlit application i have whole code in 1 main.py file which i am running by streamlit and this main.py contain all my LLM and streamlit interfaces.

---

## [crewAI and Your system has an unsupported version of sqlite3. Chroma](https://discuss.streamlit.io/t/crewai-and-your-system-has-an-unsupported-version-of-sqlite3-chroma/110493)

<div class="topic-metadata">

**Author:** [@Alireza\_K](https://discuss.streamlit.io/u/Alireza_K)\
**Replies:** 4\
**Last updated:** [May 3, 2025, 8:43pm UTC](https://discuss.streamlit.io/t/crewai-and-your-system-has-an-unsupported-version-of-sqlite3-chroma/110493 "2025-05-03T20:43:18Z")

</div>

I am using crewAI to build a very simple chatbot; it works perfectly locally but I get RuntimeError: e\[91mYour system has an unsupported version of sqlite3. Chroma requires sqlite3 \>= 3.35.0.e\[0m e\[94mPlease visit…

---

## [chatbot using ollama and openAI](https://discuss.streamlit.io/t/chatbot-using-ollama-and-openai/110228)

<div class="topic-metadata">

**Author:** [@rida4](https://discuss.streamlit.io/u/rida4)\
**Replies:** 0\
**Last updated:** [April 22, 2025, 4:35pm UTC](https://discuss.streamlit.io/t/chatbot-using-ollama-and-openai/110228 "2025-04-22T16:35:39Z")

</div>

i am trying to create a chatbot that allows you to toggle between using ollama or chatgpt, but i cannot get the latter to work. does anyone know what the issue is? if "selected\_model" not in st.session\_state: …

---

## [GPT-4-omni-mini is incredible](https://discuss.streamlit.io/t/gpt-4-omni-mini-is-incredible/75313)

<div class="topic-metadata">

**Author:** [@Sully](https://discuss.streamlit.io/u/Sully)\
**Replies:** 7\
**Last updated:** [April 7, 2025, 5:03pm UTC](https://discuss.streamlit.io/t/gpt-4-omni-mini-is-incredible/75313 "2025-04-07T17:03:37Z")

</div>

I’ve been doing a lot of testing with the new GPT-4-omni-mini model, and I’m blown away. It’s hard to believe it’s cheaper than either GPT 3.5 Turbo or Claude 3 Haiku. I updated all my toy Streamlit apps to use it, and …

---

## [{NoSessionContext} NoSessionContext() Error in StreamlitCallbackHandler while using custom agent](https://discuss.streamlit.io/t/nosessioncontext-nosessioncontext-error-in-streamlitcallbackhandler-while-using-custom-agent/59771)

<div class="topic-metadata">

**Author:** [@lz-chen](https://discuss.streamlit.io/u/lz-chen)\
**Replies:** 2\
**Last updated:** [March 10, 2025, 10:19am UTC](https://discuss.streamlit.io/t/nosessioncontext-nosessioncontext-error-in-streamlitcallbackhandler-while-using-custom-agent/59771 "2025-03-10T10:19:41Z")

</div>

Hi! I am building a chat app with streamlit and langchain. It works fine for a simple agent like this: agent = ( { "input": lambda x: x\["input"\], "agent\_scratchpad": lambda x: format\_…

---

## [Final file content rolled back to initial file content in streamlit](https://discuss.streamlit.io/t/final-file-content-rolled-back-to-initial-file-content-in-streamlit/92504)

<div class="topic-metadata">

**Author:** [@ankit8](https://discuss.streamlit.io/u/ankit8)\
**Replies:** 0\
**Last updated:** [February 8, 2025, 5:30am UTC](https://discuss.streamlit.io/t/final-file-content-rolled-back-to-initial-file-content-in-streamlit/92504 "2025-02-08T05:30:20Z")

</div>

I have created a simple streamlit application to browse files and folder and written the python code as: import streamlit as st import tkinter as tk from tkinter import filedialog import os import pandas as pd from path…

---

## [How to fix this error?](https://discuss.streamlit.io/t/how-to-fix-this-error/92233)

<div class="topic-metadata">

**Author:** [@Ankuraj](https://discuss.streamlit.io/u/Ankuraj)\
**Replies:** 1\
**Last updated:** [February 6, 2025, 6:36am UTC](https://discuss.streamlit.io/t/how-to-fix-this-error/92233 "2025-02-06T06:36:15Z")

</div>

httpx.ConnectError: This app has encountered an error. The original error message is redacted to prevent data leaks. Full error details have been recorded in the logs (if you’re on Streamlit Cloud, click on ‘Manage app’ …

---

## [Exception loading st.image as avatar](https://discuss.streamlit.io/t/exception-loading-st-image-as-avatar/91453)

<div class="topic-metadata">

**Author:** [@Gianluca1](https://discuss.streamlit.io/u/Gianluca1)\
**Replies:** 2\
**Last updated:** [January 26, 2025, 6:32am UTC](https://discuss.streamlit.io/t/exception-loading-st-image-as-avatar/91453 "2025-01-26T06:32:00Z")

</div>

Exception has occurred: StreamlitAPIException Failed to load the provided avatar value as an image. TypeError: a bytes-like object is required, not ‘DeltaGenerator’ The above exception was the direct cause of the foll…

---

## [An error occurred while parsing the content: \[Errno 111\] Connection refused](https://discuss.streamlit.io/t/an-error-occurred-while-parsing-the-content-errno-111-connection-refused/91027)

<div class="topic-metadata">

**Author:** [@Cherif](https://discuss.streamlit.io/u/Cherif)\
**Replies:** 0\
**Last updated:** [January 20, 2025, 4:09pm UTC](https://discuss.streamlit.io/t/an-error-occurred-while-parsing-the-content-errno-111-connection-refused/91027 "2025-01-20T16:09:57Z")

</div>

I deployed my app on Streamlit Cloud and it seems to be running fine, but I am getting “An error occurred while parsing the content: \[Errno 111\] Connection refused” when trying to use the LLM chat component of the app. I…

---

## [OpenAI new Realtime API](https://discuss.streamlit.io/t/openai-new-realtime-api/82594)

<div class="topic-metadata">

**Author:** [@Armin-Techlaw](https://discuss.streamlit.io/u/Armin-Techlaw)\
**Replies:** 2\
**Last updated:** [January 21, 2025, 3:05pm UTC](https://discuss.streamlit.io/t/openai-new-realtime-api/82594 "2025-01-21T15:05:24Z")

</div>

Hi there, Today OpenAI released their new API, which is multi modal GenAI. It has the famous adavnced voice mode. But it seems the API is in java and not python. Does anyone know how to use that in streamlit? Here’s th…

---

## [Error: not support 'developer' role on o1-preview and o1-mini models](https://discuss.streamlit.io/t/error-not-support-developer-role-on-o1-preview-and-o1-mini-models/90455)

<div class="topic-metadata">

**Author:** [@atsuyamaru0816](https://discuss.streamlit.io/u/atsuyamaru0816)\
**Replies:** 1\
**Last updated:** [January 14, 2025, 2:32am UTC](https://discuss.streamlit.io/t/error-not-support-developer-role-on-o1-preview-and-o1-mini-models/90455 "2025-01-14T02:32:35Z")

</div>

Hi, I’m encountering the below error when using models o1-preview and o1-mini. ■ Error Message: openai.BadRequestError: Error code: 400 - {‘error’: {‘message’: “Unsupported value: ‘messages\[0\].role’ does not support ‘d…

---

## [Plotting Lanchain REPL Code in Streamlit](https://discuss.streamlit.io/t/plotting-lanchain-repl-code-in-streamlit/90435)

<div class="topic-metadata">

**Author:** [@\_Saad](https://discuss.streamlit.io/u/_Saad)\
**Replies:** 0\
**Last updated:** [January 13, 2025, 7:41pm UTC](https://discuss.streamlit.io/t/plotting-lanchain-repl-code-in-streamlit/90435 "2025-01-13T19:41:54Z")

</div>

Hi there, I am building a webapp utilizing Replit AI Agent, and streamlit has been of great help throughout. However, I have been stuck on one specific situation for the past three days that I just can’t get past. S…

---

## [How to Stream Messages with Assistant Icon?](https://discuss.streamlit.io/t/how-to-stream-messages-with-assistant-icon/90422)

<div class="topic-metadata">

**Author:** [@RP1](https://discuss.streamlit.io/u/RP1)\
**Replies:** 0\
**Last updated:** [January 13, 2025, 3:58pm UTC](https://discuss.streamlit.io/t/how-to-stream-messages-with-assistant-icon/90422 "2025-01-13T15:58:08Z")

</div>

Currently the message streams in a placeholder, disappears, then reappears with assistant icon. I want the message to stream WITH the assistant chat bubble from the start. # Current code message\_placeholder = st.empty()…

---

## [Health check to check for exception?](https://discuss.streamlit.io/t/health-check-to-check-for-exception/90190)

<div class="topic-metadata">

**Author:** [@Joe6](https://discuss.streamlit.io/u/Joe6)\
**Replies:** 3\
**Last updated:** [January 10, 2025, 8:52pm UTC](https://discuss.streamlit.io/t/health-check-to-check-for-exception/90190 "2025-01-10T20:52:54Z")

</div>

Is there a simple way to see if the streamlit code passes without exception? Seems like binary xhr’s are passed and having selenium call script locally is slow and feels like overkill… Just want to get back what except…

---

## [Menu page calling a script that runs langchain](https://discuss.streamlit.io/t/menu-page-calling-a-script-that-runs-langchain/88790)

<div class="topic-metadata">

**Author:** [@lmoliveira](https://discuss.streamlit.io/u/lmoliveira)\
**Replies:** 0\
**Last updated:** [January 3, 2025, 6:38pm UTC](https://discuss.streamlit.io/t/menu-page-calling-a-script-that-runs-langchain/88790 "2025-01-03T18:38:50Z")

</div>

Dear, I have a chatbot script that runs perfectly from the terminal. This script runs a = qa\_chain.invoke({“query”: query\]}) b =a\[“result”\] st.write(b) But when I call it from the home.py, it runs everything until t…

[Next page](https://discuss.streamlit.io/c/llm/26.md?page=1)
