Improve Qwen streamlit error handling#104

Open

jeonghoonkang wants to merge 1 commit intomasterfrom

km7msi-codex/add-execution-error-output-window

Owner

jeonghoonkang commented Jul 26, 2025

Summary

show GPU memory usage and model max tokens in Qwen Streamlit demo
mention new info in README
rely on transformers automatic device placement when loading the model
allow selecting the GGUF 1.8B model and add emoji markers for Q&A

Testing

pytest -q (fails: missing dependencies)

https://chatgpt.com/codex/tasks/task_e_6884095f16208331b972a62cbfdfa275


          Add GGUF model option and emoji UI

310ff6a

jeonghoonkang added the codex label

— with

ChatGPT Codex Connector

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels