Closed ApexArray closed 1 week ago
Thanks for the heads up, glad to hear its helped you!
I've gone ahead and added that feature.
Right now I'm focused on getting to v1.0 before writing up a CONTRIBUTING.md
file, but if you'd like to open a PR for something please do, I promise I'll at least look at/review it.
Hello! Great work on this project so far! It's already saved me several hours of summarizing training videos and taking notes.
Is your feature request related to a problem? Please describe. Ollama summarizations frequently fail due to the hard-coded 30s timeout.
Describe the solution you'd like Add a config option
summarize_api_timeout
(or similar) to allow setting a custom timeout value.Describe alternatives you've considered Remove the timeout all together (other APIs like llama.cpp don't seem to have a timeout configured).
Additional context Here's my Ollama output for a summarization chunking 500 words (100 overlap) on a GTX 1080ti. Notice that the first request failed with an
HTTP 500
error after 30 seconds.Also, I'm willing to help with development if you're looking for contributors. Let me know if you'd like me to open a PR.