Fix logic for default modelMaxTokens for thinking models #1476
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Context
@ReadyPlayerEmma noticed that even though the thinking models visually default to 16k max tokens in the case where the user hasn't touched the max tokens slider, they were being treated as 128k max tokens in the sliding window calculations.
Implementation
This is the quickest fix of just hardcoding to 16k if undefined like we do elsewhere. Really need to clean this code up though.
Important
Fixes default
modelMaxTokensfor thinking models to 16,384 inCline.tsand updates tests to verify this behavior.modelMaxTokensfor thinking models inCline.tsto 16,384 if undefined.DEFAULT_THINKING_MODEL_MAX_TOKENSinmodel-utils.tsset to 16,384.getMaxTokensForModel.test.tsxto test fallback toDEFAULT_THINKING_MODEL_MAX_TOKENS.model-utils.test.tsto verify behavior for thinking models with undefinedmodelMaxTokens.This description was created by
for da286c4. It will automatically update as commits are pushed.