Amazon Nova 2 Omni
nova-2-omni
ProprietaryMultimodal
About
Amazon Nova 2 Omni by AWS. The industry's first multimodal reasoning model supporting text, images, video, and speech inputs while generating both text and image outputs. Announced at AWS re:Invent 2025 (December 2). Supports 1M token context, 200+ languages, and native image generation with character consistency and text rendering. Available in preview via Amazon Bedrock (aws/nova-2-omni).
Amazon Nova 2 Omni has a 1M-token context window.
Capabilities
VisionMultimodalReasoningFunction CallingTool UseStructured OutputsCode ExecutionPrompt CachingBatch APIAudioFine-tuning
Specifications
FamilyAmazon Nova
Released2025-12-02
Context1M
Specializationmultimodal
LicenseProprietary
