3 ms·
Interesting. I'm on Jetbrains+Copilot and I find that the autocomplete is useful 90% of the time. I'm using the Claude backend instead of the default (OpenAI).
by selcuka 2y ago
Interesting. I'm on Jetbrains+Copilot and I find that the autocomplete is useful 90% of the time.
I'm using the Claude backend instead of the default (OpenAI). I wonder if that makes the difference.
- aerhardt 2y agoI only have the 4o option for completions. Like, I see the option in the settings but the dropdown only produces 4o for me. I seem to remember that in github.com I have all the models exposed, but in PyCharm I only get that for autocomplete. I do get a wider model choice in the chat pane. I'll look into it.
- selcuka 2y agoI stand corrected. Claude is enabled on my GitHub.com settings but it's not yet available to PyCharm, so I'm using GPT-4o as well. It's a known issue: https://github.com/orgs/community/discussions/143139 https://github.com/orgs/community/discussions/143139