Sorting LiteLLM GitHub issues by difficulty to implement. Higher = harder.

{
#26326 requires coordinated streaming and non-streaming translation changes, stateful tag parsing across chunk boundaries, provider-specific handler behavior, and compatibility tests. #34228 appears more localized to response-shape normalization, though it may require preserving multiple turns and tool-call semantics. The former carries greater cross-path regression risk.
}
https://github.com/berriai/litellm/issues/26326 3:1 https://github.com/berriai/litellm/issues/34228