fix(ai): enable Bedrock prompt caching for Claude 5

closes #6235

packages/ai/CHANGELOG.md

@@ -7,6 +7,7 @@
 - Fixed GitHub Copilot device-code login polling to wait before the first token poll, avoiding incorrect device-code failures for some users after browser authorization ([#6187](https://github.com/earendil-works/pi/issues/6187)).
 - Fixed OpenAI Codex user-agent construction to synchronously load Node OS metadata, avoiding a startup race that could report `pi (browser)` in Node/Bun.
 - Fixed Fireworks GLM 5.2 Fast to use the OpenAI-compatible endpoint and `thinkingLevelMap`, aligning it with GLM 5.2 ([#6195](https://github.com/earendil-works/pi/issues/6195)).
+- Fixed Amazon Bedrock prompt-cache points for Claude Fable 5 and Claude Sonnet 5 ([#6235](https://github.com/earendil-works/pi/issues/6235)).
 
 ### Added
 

packages/ai/src/api/bedrock-converse-stream.ts

@@ -640,7 +640,7 @@ function isAnthropicClaudeModel(model: Model<"bedrock-converse-stream">): boolea
 
 /**
  * Check if the model supports prompt caching.
- * Supported: Claude 3.5 Haiku, Claude 3.7 Sonnet, Claude 4.x models
+ * Supported: Claude 3.5 Haiku, Claude 3.7 Sonnet, Claude 4.x models, Claude 5 models
  *
  * For base models and system-defined inference profiles the model ID / ARN
  * contains the model name, so we can decide locally.
@@ -660,6 +660,8 @@ function supportsPromptCaching(model: Model<"bedrock-converse-stream">, env?: Pr
 		if (getProviderEnvValue("AWS_BEDROCK_FORCE_CACHE", env) === "1") return true;
 		return false;
 	}
+	// Claude 5 models (fable-5, sonnet-5)
+	if (candidates.some((s) => s.includes("fable-5") || s.includes("sonnet-5"))) return true;
 	// Claude 4.x models (opus-4, sonnet-4, haiku-4)
 	if (candidates.some((s) => s.includes("-4-"))) return true;
 	// Claude 3.7 Sonnet