fix: remove explicit cache_control for Google models in OpenRouter (#4487)

- Remove all Google models from OPEN_ROUTER_PROMPT_CACHING_MODELS set
- This resolves 3+ minute lag when using google/gemini-2.5-pro-preview
- OpenRouter still provides automatic implicit ephemeral caching for these models
- Updated tests to handle intentional exclusion of Google models from explicit caching

Fixes #4487

Author: hannesrudolph

packages/types/src/providers/openrouter.ts

@@ -39,14 +39,6 @@ export const OPEN_ROUTER_PROMPT_CACHING_MODELS = new Set([
 	"anthropic/claude-3.7-sonnet:thinking",
 	"anthropic/claude-sonnet-4",
 	"anthropic/claude-opus-4",
-	"google/gemini-2.5-pro-preview",
-	"google/gemini-2.5-flash-preview",
-	"google/gemini-2.5-flash-preview:thinking",
-	"google/gemini-2.5-flash-preview-05-20",
-	"google/gemini-2.5-flash-preview-05-20:thinking",
-	"google/gemini-2.0-flash-001",
-	"google/gemini-flash-1.5",
-	"google/gemini-flash-1.5-8b",
 ])
 
 // https://www.anthropic.com/news/3-5-models-and-computer-use

src/api/providers/fetchers/__tests__/openrouter.spec.ts

@@ -23,12 +23,37 @@ describe("OpenRouter API", () => {
 
 			const models = await getOpenRouterModels()
 
-			expect(
-				Object.entries(models)
-					.filter(([_, model]) => model.supportsPromptCache)
-					.map(([id, _]) => id)
-					.sort(),
-			).toEqual(Array.from(OPEN_ROUTER_PROMPT_CACHING_MODELS).sort())
+			// Check that our caching models list is a subset of what OpenRouter supports
+			// (we may intentionally exclude some models that support caching)
+			const openRouterSupportedCaching = Object.entries(models)
+				.filter(([_, model]) => model.supportsPromptCache)
+				.map(([id, _]) => id)
+
+			const ourCachingModels = Array.from(OPEN_ROUTER_PROMPT_CACHING_MODELS)
+
+			// Verify all our caching models are actually supported by OpenRouter
+			for (const modelId of ourCachingModels) {
+				expect(openRouterSupportedCaching).toContain(modelId)
+			}
+
+			// Check that we have the expected models (excluding intentionally excluded ones)
+			const excludedCachingModels = new Set([
+				// All Google models intentionally excluded from caching
+				"google/gemini-2.5-pro-preview",
+				"google/gemini-2.0-flash-001",
+				"google/gemini-2.5-flash-preview",
+				"google/gemini-2.5-flash-preview-05-20",
+				"google/gemini-2.5-flash-preview-05-20:thinking",
+				"google/gemini-2.5-flash-preview:thinking",
+				"google/gemini-flash-1.5",
+				"google/gemini-flash-1.5-8b",
+			])
+
+			const expectedCachingModels = openRouterSupportedCaching
+				.filter((id) => !excludedCachingModels.has(id))
+				.sort()
+
+			expect(ourCachingModels.sort()).toEqual(expectedCachingModels)
 
 			expect(
 				Object.entries(models)