Skip to content

[BOT ISSUE] Vertex: add missing cache read pricing for Gemini 2.5 Flash and Pro #1079

Description

@github-actions

Vertex Gemini 2.5 entries missing cache read pricing

The Vertex-path entries for Gemini 2.5 Flash and Gemini 2.5 Pro are missing input_cache_read_cost_per_mil_tokens, while their Google AI Studio counterparts have the correct values.

Affected models

Model ID (Vertex) Missing Field Correct Value Google AI Studio entry
publishers/google/models/gemini-2.5-flash input_cache_read_cost_per_mil_tokens 0.03 gemini-2.5-flash has 0.03
publishers/google/models/gemini-2.5-pro input_cache_read_cost_per_mil_tokens 0.125 gemini-2.5-pro has 0.125

The Google AI Studio entries (gemini-2.5-flash, gemini-2.5-pro) correctly include cache read pricing. The Vertex entries represent the same models but are missing this field, causing inconsistent cost calculations depending on which entry is used.

Sources

  • Catalog internal comparison: gemini-2.5-flash (line 11559) vs publishers/google/models/gemini-2.5-flash (line 6505)
  • Vertex AI Pricing — confirms context caching pricing for both models

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions