AWS bedrock documentation change
Summary
Added support for Claude Opus 4.7 and Claude Mythos Preview models, clarified deprecation of manual extended thinking, and corrected parameter placement for 'effort' setting with API examples
Security assessment
The changes focus on feature updates (new model support), deprecation notices for API parameters, and documentation corrections for proper parameter usage. There's no evidence of security vulnerability fixes, incident response, or security feature additions. The 400 error mentioned is a standard validation error for incorrect API usage.
Diff
diff --git a/bedrock/latest/userguide/claude-messages-adaptive-thinking.md b/bedrock/latest/userguide/claude-messages-adaptive-thinking.md index c94d7cc51..c7fe66d4b 100644 --- a//bedrock/latest/userguide/claude-messages-adaptive-thinking.md +++ b//bedrock/latest/userguide/claude-messages-adaptive-thinking.md @@ -16,0 +17,2 @@ Model | Model ID +Claude Opus 4.7 | `anthropic.claude-opus-4-7` +Claude Mythos Preview | `anthropic.claude-mythos-preview` @@ -22 +24,3 @@ Claude Sonnet 4.6 | `anthropic.claude-sonnet-4-6` -`thinking.type: "enabled"` and `budget_tokens` are deprecated on Claude Opus 4.6 and will be removed in a future model release. Use `thinking.type: "adaptive"` with the effort parameter instead. +Claude Opus 4.7 and Claude Mythos Preview _only_ support adaptive thinking. Manual extended thinking (`thinking.type: "enabled"` with `budget_tokens`) is not supported on these models and will return a 400 error. + +`thinking.type: "enabled"` and `budget_tokens` are deprecated on Claude Opus 4.6 and Claude Sonnet 4.6 and will be removed in a future model release. Use `thinking.type: "adaptive"` with the effort parameter instead. @@ -139,0 +144,22 @@ Effort level | Thinking behavior +###### Important + +The `effort` parameter must be placed inside a separate `output_config` object in your request body — not inside the `thinking` object. Placing `effort` inside `thinking` will result in a `ValidationException`. + +The following example shows how to set the effort level when using the InvokeModel API: + + + { + "anthropic_version": "bedrock-2023-05-31", + "max_tokens": 16000, + "thinking": { + "type": "adaptive" + }, + "output_config": { + "effort": "high" + }, + "messages": [{ + "role": "user", + "content": "Your prompt here" + }] + } + @@ -164 +190 @@ When using the [Converse API](./conversation-inference.html), pass the `thinking -To specify an effort level, add the `effort` field inside the `thinking` object: +To specify an effort level, add the `effort` field inside a separate `output_config` object in `additionalModelRequestFields`: @@ -175 +201,3 @@ To specify an effort level, add the `effort` field inside the `thinking` object: - "type": "adaptive", + "type": "adaptive" + }, + "output_config": {