AWS bedrock: Nova 2.5 Sonic model card: icon alt-text cleanup and watermarking support
Summary
Replaces verbose image alt text ("Red circle with white X icon..." / "Green circle with white checkmark icon...") with concise "not-supported"/"supported" alt text throughout the model card's capability, feature, tier, and region tables, and adds a new supported-feature bullet for Watermarking.
Security assessment
The diff is an accessibility/alt-text refactor of support-matrix icons plus one added capability bullet noting that Watermarking is supported for this model. Watermarking relates to content provenance/authenticity rather than a vulnerability fix, and no CVE, credential, IAM, encryption, or exposure change is present.
Evidence
+ *  [Watermarking](watermarking.html)
Diff
diff --git a/bedrock/latest/userguide/model-card-amazon-nova-2-5-sonic.md b/bedrock/latest/userguide/model-card-amazon-nova-2-5-sonic.md index 9ebdd57..c583dc3 100644 --- a//bedrock/latest/userguide/model-card-amazon-nova-2-5-sonic.md +++ b//bedrock/latest/userguide/model-card-amazon-nova-2-5-sonic.md @@ -40,5 +40,5 @@ Nova 2.5 Sonic is Amazon's latest speech-to-speech foundation model for building - Audio|  Embedding|  `Responses`|  `bedrock-runtime` - Image|  Image|  `Chat Completions`|  `bedrock-mantle` - Speech|  Speech|  `Invoke`| - Text|  Text|  `Converse`| - Video|  Video|  `InvokeModelWithBidirectionalStream`| + Audio|  Embedding|  `Responses`|  `bedrock-runtime` + Image|  Image|  `Chat Completions`|  `bedrock-mantle` + Speech|  Speech|  `Invoke`| + Text|  Text|  `Converse`| + Video|  Video|  `InvokeModelWithBidirectionalStream`| @@ -59,3 +59,4 @@ Whenever possible, we recommend using the `bedrock-runtime` endpoint for new app - *  [Response streaming](/bedrock/latest/APIReference/API_runtime_InvokeModelWithResponseStream.html) - *  Tool use (function calling) - *  Asynchronous tool calling + *  [Response streaming](/bedrock/latest/APIReference/API_runtime_InvokeModelWithResponseStream.html) + *  Tool use (function calling) + *  Asynchronous tool calling + *  [Watermarking](watermarking.html) @@ -65,10 +66,10 @@ Whenever possible, we recommend using the `bedrock-runtime` endpoint for new app - *  [Intelligent prompt routing](prompt-routing.html) - *  [Abuse detection](abuse-detection.html) - *  [Guardrails](guardrails.html) - *  [Prompt optimization](prompt-management-optimize.html) - *  [Count tokens](count-tokens.html) - *  [Knowledge base](knowledge-base.html) - *  [Model evaluation](evaluation.html) - *  [Prompt management](prompt-management.html) - *  [Flows](flows.html) - *  [Agents](agents.html) + *  [Intelligent prompt routing](prompt-routing.html) + *  [Abuse detection](abuse-detection.html) + *  [Guardrails](guardrails.html) + *  [Prompt optimization](prompt-management-optimize.html) + *  [Count tokens](count-tokens.html) + *  [Knowledge base](knowledge-base.html) + *  [Model evaluation](evaluation.html) + *  [Prompt management](prompt-management.html) + *  [Flows](flows.html) + *  [Agents](agents.html) @@ -80 +81 @@ Whenever possible, we recommend using the `bedrock-runtime` endpoint for new app - *  **Speech-to-speech** — Real-time bidirectional speech understanding and generation in a single model. + *  **Speech-to-speech** — Real-time bidirectional speech understanding and generation in a single model. @@ -82 +83 @@ Whenever possible, we recommend using the `bedrock-runtime` endpoint for new app - *  **Enhanced reasoning** — Stronger multi-step reasoning during live conversations. + *  **Enhanced reasoning** — Stronger multi-step reasoning during live conversations. @@ -84 +85 @@ Whenever possible, we recommend using the `bedrock-runtime` endpoint for new app - *  **Improved instruction following** — More reliable adherence to system prompts, persona instructions, and multi-step directives. + *  **Improved instruction following** — More reliable adherence to system prompts, persona instructions, and multi-step directives. @@ -86 +87 @@ Whenever possible, we recommend using the `bedrock-runtime` endpoint for new app - *  **Tool use (function calling)** — Higher-accuracy tool calling for agentic workflows: look up an order, check eligibility, and take action in one session. + *  **Tool use (function calling)** — Higher-accuracy tool calling for agentic workflows: look up an order, check eligibility, and take action in one session. @@ -88 +89 @@ Whenever possible, we recommend using the `bedrock-runtime` endpoint for new app - *  **Asynchronous tool calling** — Keeps the conversation going while tools run in the background. + *  **Asynchronous tool calling** — Keeps the conversation going while tools run in the background. @@ -90 +91 @@ Whenever possible, we recommend using the `bedrock-runtime` endpoint for new app - *  **Cross-modal input** — Mix text and voice in the same session. + *  **Cross-modal input** — Mix text and voice in the same session. @@ -92 +93 @@ Whenever possible, we recommend using the `bedrock-runtime` endpoint for new app - *  **Polyglot voices** — Voices that speak naturally across multiple languages. + *  **Polyglot voices** — Voices that speak naturally across multiple languages. @@ -94 +95 @@ Whenever possible, we recommend using the `bedrock-runtime` endpoint for new app - *  **Natural turn-taking** — Graceful handling of user interruptions with configurable turn detection. + *  **Natural turn-taking** — Graceful handling of user interruptions with configurable turn detection. @@ -96 +97 @@ Whenever possible, we recommend using the `bedrock-runtime` endpoint for new app - *  **Code-switching** — Switches between languages mid-conversation. + *  **Code-switching** — Switches between languages mid-conversation. @@ -98 +99 @@ Whenever possible, we recommend using the `bedrock-runtime` endpoint for new app - *  **Streaming text transcripts** — Real-time text transcription of the voice conversation. + *  **Streaming text transcripts** — Real-time text transcription of the voice conversation. @@ -100 +101 @@ Whenever possible, we recommend using the `bedrock-runtime` endpoint for new app - *  **Lower latency** — Reduced response times for more natural conversations. + *  **Lower latency** — Reduced response times for more natural conversations. @@ -125 +126 @@ Amazon Bedrock offers multiple service tiers to match your workload requirements - |  |  |  + |  |  |  @@ -135,4 +136,4 @@ Amazon Bedrock offers three inference options: **In-Region** keeps requests with -`us-east-1` (N. Virginia)| | |  -`us-west-2` (Oregon)| | |  -`eu-north-1` (Stockholm)| | |  -`ap-northeast-1` (Tokyo)| | |  +`us-east-1` (N. Virginia)| | |  +`us-west-2` (Oregon)| | |  +`eu-north-1` (Stockholm)| | |  +`ap-northeast-1` (Tokyo)| | | 