mirror of
https://github.com/BerriAI/litellm.git
synced 2026-10-06 02:48:13 +00:00
322ae4e749
Vertex charges a 10% premium for Claude models served from regional and multi-region endpoints, while our cost map only carried the global endpoint rates. Any deployment pinned to a location such as us-east5 was therefore logged 10% under what Vertex bills. Adds regional_endpoint_uplift_multiplier to the cost map for the Claude models Google prices that way, threads the deployment's vertex_location into the Vertex cost calculator, and applies the multiplier to every token type, including the cache write and cache read rates. A location of global, or no location at all, keeps the previous pricing. Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> |
||
|---|---|---|
| .. | ||
| baseline_db.py | ||
| check_file_length.py | ||
| check_files_match.py | ||
| generate_model_prices_schema.py | ||
| run_migration.py | ||
| security_scans_readme.md | ||
| TEST_KEY_PATTERNS.md | ||