feat(aws-bedrock): add new models [bot]

[hganwani-droid]

Auto-generated by model-addition-agent for provider aws-bedrock.

---

Note

Medium Risk
Primarily data/config updates, but changing modalities and region pricing metadata can affect model selection, capability gating (e.g., text/image/audio), and any downstream cost calculations.

Overview
AWS Bedrock model registry updates. Adds/normalizes modalities (explicit input/output types like text, image, audio) across a large set of Bedrock model YAMLs, including chat, embedding, rerank, image, and video models.

Also expands pricing/availability metadata by adding additional costs regions for multiple models/inference profiles (e.g., new ap-southeast-*, eu-*, me-*, ca-* entries), with a few minor adjustments to existing modality lists (e.g., embeddings now include text input and text output where missing).

^{Written by Cursor Bugbot for commit 53d3a4d. This will update automatically on new commits. Configure here.}

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

^{Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.}

[cursor]

Audio modalities removed from sonic speech models

High Severity

The audio modality was removed from both Nova Sonic models, and the entire input modalities section was deleted. These are speech-to-speech models — their costs include input_cost_per_audio_token and output_cost_per_audio_token in every region, confirming audio is a core capability. The modalities now only list output: text, which misrepresents these audio-first models.

Additional Locations (1)

• providers/aws-bedrock/amazon.nova-sonic-v1:0.yaml#L22-L25