Add Gemma 4 models to bedrock-mantle

[akhilmmenon]

Summary

• Add google.gemma-4-31b, google.gemma-4-26b-a4b, and google.gemma-4-e2b to general/bedrock-mantle.json and pricing/bedrock-mantle.json.
• Pricing uses AWS Bedrock US On-Demand rates (same convention as existing Gemma 3 entries).
• max_tokens cap set to 32768, matching other bedrock-mantle chat models.

Pylon: #6853 — Enabling Gemma 4 on PortKey (AWS Bedrock Mantle)

Test plan

• jq empty general/bedrock-mantle.json pricing/bedrock-mantle.json
• Verify models appear in catalog after S3 sync
• Smoke test chat completion via bedrock-mantle provider after deploy

[Copilot]

Pull request overview

Adds newly available Google Gemma 4 models to the AWS Bedrock Mantle provider catalog so they can be selected and costed consistently alongside existing Bedrock Mantle chat models.

Changes:

• Added three Gemma 4 model identifiers to general/bedrock-mantle.json with max_tokens capped at 32768 and multimodal/tool capability flags.
• Added corresponding pay-as-you-go token pricing entries to pricing/bedrock-mantle.json, following the existing Bedrock Mantle Gemma pricing structure.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.

File	Description
general/bedrock-mantle.json	Registers google.gemma-4-* models in the Bedrock Mantle model catalog with token limits and supported capabilities.
pricing/bedrock-mantle.json	Defines input/output token prices for the new Gemma 4 models under Bedrock Mantle pay-as-you-go pricing.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.