AWS bedrock documentation change
Summary
Added new Claude Mythos 5 and Claude Fable 5 models, introduced 'Billing notes' section explaining refusal billing behavior
Security assessment
Billing clarification for refused requests describes accounting behavior rather than security features. New model introductions don't include security-specific documentation.
Diff
diff --git a/bedrock/latest/userguide/model-cards-anthropic.md b/bedrock/latest/userguide/model-cards-anthropic.md index f9f372a6e..07df0cab8 100644 --- a//bedrock/latest/userguide/model-cards-anthropic.md +++ b//bedrock/latest/userguide/model-cards-anthropic.md @@ -6,0 +7,2 @@ +Billing notes + @@ -12,0 +15,2 @@ The following Anthropic models are available in Amazon Bedrock: +[Claude Mythos 5](./model-card-anthropic-claude-mythos-5.html)| Claude Mythos 5 is Anthropic's most capable model for cybersecurity and life sciences, including vulnerability discovery, drug design, and biodefense screening. +[Claude Fable 5](./model-card-anthropic-claude-fable-5.html)| Claude Fable 5 is Anthropic's next-generation model for complex knowledge work and coding, capable of sustained autonomous operation across multi-day tasks. @@ -25,0 +30,9 @@ The following Anthropic models are available in Amazon Bedrock: +## Billing notes + +The following billing behaviors apply specifically to Anthropic models on Amazon Bedrock: + + * **Refused requests (all Anthropic models on`bedrock-mantle`):** When a request is refused by a content classifier before inference begins, no input or output tokens are billed. Mid-stream refusals — where the classifier fires after output has begun streaming — are billed for tokens consumed before the block. Both refusal types return `stop_reason: "refusal"` in the response. + + + + @@ -34 +47 @@ Titan Embeddings G1 - Text v2 -Claude Opus 4.8 +Claude Mythos 5