fix(gemini): forward OpenAI usage-only chunks and cached/thought token counts to /v1beta clients - #17842

fix(gemini): forward OpenAI usage-only chunks and cached/thought token counts to /v1beta clients#17842

diegosouzapw merged 3 commits into

cheaper-inference:release/v3.8.53cheaper-inference/OmniRoute:release/v3.8.53from

AndresDiagoM:fix/gemini-usage-metadata-from-openaiAndresDiagoM/OmniRoute:fix/gemini-usage-metadata-from-openaiCopy head branch name to clipboard