Dallin Romney

@Symfomany have you tried it? Is it fixed for you on 1.0.21+?

@0x23d11 before writing tests for this note that https://github.com/continuedev/continue/pull/5138 changes pruning logic quite a bit and will effect this. EDIT: didn't look closely enough, looks like this is for pruning...

@0x23d11 There are some tests in countTokens.test.ts that you could just flesh out a bit with more examples and then unskip!

Great, will take a look!

@ferenci84 I took forever to follow up on this so I went ahead and touched up. Appreciate the contribution!

@akulbector this is super helpful, thank you for the in-depth digging. It's possible that the new `reponses` API support for GPT-5 models somehow broke access through openrouter by attempting to...

Yes, #8417 would have been it! Noting this line of code: ``` const RESPONSES_MODEL_REGEX = /^(?:gpt-5|gpt-5-codex|o)/i; export function isResponsesModel(model: string): boolean { return !!model && RESPONSES_MODEL_REGEX.test(model); } ``` https://github.com/continuedev/continue/pull/8417/files#diff-fa3d4d4922c9bd1f3cd2a1decd074b3e47cb961816e88e19256c359e5dec65c4 Could...

@alanwilter using your prompt and function definition with gpt-4o it worked for me. Can you confirm that your pytest.prompt file was properly saved? /.prompts/pytest.prompt ```mdx You are an expert programmer...

@alanwilter agree, switched to prerelease while keeping model prompt etc. the same and experienced same thing.

Could approach as "pinned chat" where you can pin a specific chat to the top at any time, unpin at any time, and pin icon or similar + keyboard shortcut...