[refactor] Improve backend service URL handling and logging#2246

Open

charlieyl wants to merge 24 commits intodev/v0.7.0from charlie/dev/v0.7.0

+329-114

Commits on Dec 28, 2024

[logs]add deploy param
charlieyl
committed

Commits on Feb 11, 2025

[bugfix] Handle deployment failure by deleting deployed replicas and releasing GPU
charlieyl
committed

Commits on Feb 12, 2025

Commits on Feb 13, 2025

[perf] Remove verbose logging in model inference request handling
charlieyl
committed
[bugfix-combination] Add model configuration details for deployment failure handling
charlieyl
committed
[bugfix] Update default model configuration parameters for safer deployment
charlieyl
committed

Commits on Feb 14, 2025

[feature] Add endpoint_name parameter to model deployment method
charlieyl
committed

Commits on Feb 17, 2025

[bugfix] Restore full GPU card selection parameters in NvidiaGPUtil
charlieyl
committed