diff --git a/packages/types/src/providers/bedrock.ts b/packages/types/src/providers/bedrock.ts
index 9935d90b127a..10676dd19d98 100644
--- a/packages/types/src/providers/bedrock.ts
+++ b/packages/types/src/providers/bedrock.ts
@@ -26,6 +26,15 @@ export const bedrockModels = {
 		minTokensPerCachePoint: 1024,
 		maxCachePoints: 4,
 		cachableFields: ["system", "messages", "tools"],
+		tiers: [
+			{
+				contextWindow: 1_000_000,
+				inputPrice: 6.0,
+				outputPrice: 22.5,
+				cacheWritesPrice: 7.5,
+				cacheReadsPrice: 0.6,
+			},
+		],
 	},
 	"amazon.nova-pro-v1:0": {
 		maxTokens: 5000,
@@ -90,6 +99,15 @@ export const bedrockModels = {
 		minTokensPerCachePoint: 1024,
 		maxCachePoints: 4,
 		cachableFields: ["system", "messages", "tools"],
+		tiers: [
+			{
+				contextWindow: 1_000_000,
+				inputPrice: 6.0,
+				outputPrice: 22.5,
+				cacheWritesPrice: 7.5,
+				cacheReadsPrice: 0.6,
+			},
+		],
 	},
 	"anthropic.claude-opus-4-1-20250805-v1:0": {
 		maxTokens: 8192,
diff --git a/packages/types/src/providers/gemini.ts b/packages/types/src/providers/gemini.ts
index aae428d90c02..c8bac1f5b30c 100644
--- a/packages/types/src/providers/gemini.ts
+++ b/packages/types/src/providers/gemini.ts
@@ -138,24 +138,19 @@ export const geminiModels = {
 		contextWindow: 1_048_576,
 		supportsImages: true,
 		supportsPromptCache: true,
-		inputPrice: 2.5, // This is the pricing for prompts above 200k tokens.
-		outputPrice: 15,
-		cacheReadsPrice: 0.625,
-		cacheWritesPrice: 4.5,
+		inputPrice: 1.25, // Base price for ≤200k tokens
+		outputPrice: 10,
+		cacheReadsPrice: 0.125,
+		cacheWritesPrice: 1.625,
 		maxThinkingTokens: 32_768,
 		supportsReasoningBudget: true,
 		tiers: [
 			{
-				contextWindow: 200_000,
-				inputPrice: 1.25,
-				outputPrice: 10,
-				cacheReadsPrice: 0.31,
-			},
-			{
-				contextWindow: Infinity,
-				inputPrice: 2.5,
+				contextWindow: 1_000_000,
+				inputPrice: 2.5, // >200k tokens
 				outputPrice: 15,
-				cacheReadsPrice: 0.625,
+				cacheReadsPrice: 0.25,
+				cacheWritesPrice: 2.875,
 			},
 		],
 	},
@@ -164,22 +159,17 @@ export const geminiModels = {
 		contextWindow: 1_048_576,
 		supportsImages: true,
 		supportsPromptCache: true,
-		inputPrice: 2.5, // This is the pricing for prompts above 200k tokens.
-		outputPrice: 15,
-		cacheReadsPrice: 0.625,
-		cacheWritesPrice: 4.5,
+		inputPrice: 1.25, // Base price for ≤200k tokens
+		outputPrice: 10,
+		cacheReadsPrice: 0.125,
+		cacheWritesPrice: 1.625,
 		tiers: [
 			{
-				contextWindow: 200_000,
-				inputPrice: 1.25,
-				outputPrice: 10,
-				cacheReadsPrice: 0.31,
-			},
-			{
-				contextWindow: Infinity,
-				inputPrice: 2.5,
+				contextWindow: 1_000_000,
+				inputPrice: 2.5, // >200k tokens
 				outputPrice: 15,
-				cacheReadsPrice: 0.625,
+				cacheReadsPrice: 0.25,
+				cacheWritesPrice: 2.875,
 			},
 		],
 	},
@@ -188,24 +178,19 @@ export const geminiModels = {
 		contextWindow: 1_048_576,
 		supportsImages: true,
 		supportsPromptCache: true,
-		inputPrice: 2.5, // This is the pricing for prompts above 200k tokens.
-		outputPrice: 15,
-		cacheReadsPrice: 0.625,
-		cacheWritesPrice: 4.5,
+		inputPrice: 1.25, // Base price for ≤200k tokens
+		outputPrice: 10,
+		cacheReadsPrice: 0.125,
+		cacheWritesPrice: 1.625,
 		maxThinkingTokens: 32_768,
 		supportsReasoningBudget: true,
 		tiers: [
 			{
-				contextWindow: 200_000,
-				inputPrice: 1.25,
-				outputPrice: 10,
-				cacheReadsPrice: 0.31,
-			},
-			{
-				contextWindow: Infinity,
-				inputPrice: 2.5,
+				contextWindow: 1_000_000,
+				inputPrice: 2.5, // >200k tokens
 				outputPrice: 15,
-				cacheReadsPrice: 0.625,
+				cacheReadsPrice: 0.25,
+				cacheWritesPrice: 2.875,
 			},
 		],
 	},
@@ -222,25 +207,20 @@ export const geminiModels = {
 		contextWindow: 1_048_576,
 		supportsImages: true,
 		supportsPromptCache: true,
-		inputPrice: 2.5, // This is the pricing for prompts above 200k tokens.
-		outputPrice: 15,
-		cacheReadsPrice: 0.625,
-		cacheWritesPrice: 4.5,
+		inputPrice: 1.25, // Base price for ≤200k tokens
+		outputPrice: 10,
+		cacheReadsPrice: 0.125,
+		cacheWritesPrice: 1.625,
 		maxThinkingTokens: 32_768,
 		supportsReasoningBudget: true,
 		requiredReasoningBudget: true,
 		tiers: [
 			{
-				contextWindow: 200_000,
-				inputPrice: 1.25,
-				outputPrice: 10,
-				cacheReadsPrice: 0.31,
-			},
-			{
-				contextWindow: Infinity,
-				inputPrice: 2.5,
+				contextWindow: 1_000_000,
+				inputPrice: 2.5, // >200k tokens
 				outputPrice: 15,
-				cacheReadsPrice: 0.625,
+				cacheReadsPrice: 0.25,
+				cacheWritesPrice: 2.875,
 			},
 		],
 	},
diff --git a/packages/types/src/providers/vertex.ts b/packages/types/src/providers/vertex.ts
index f277c58a3ef2..d15d583ef1cd 100644
--- a/packages/types/src/providers/vertex.ts
+++ b/packages/types/src/providers/vertex.ts
@@ -61,49 +61,79 @@ export const vertexModels = {
 		contextWindow: 1_048_576,
 		supportsImages: true,
 		supportsPromptCache: true,
-		inputPrice: 2.5,
-		outputPrice: 15,
+		inputPrice: 1.25,
+		outputPrice: 10,
+		cacheReadsPrice: 0.125,
+		cacheWritesPrice: 1.625,
+		tiers: [
+			{
+				contextWindow: 1_000_000,
+				inputPrice: 2.5,
+				outputPrice: 15,
+				cacheReadsPrice: 0.25,
+				cacheWritesPrice: 2.875,
+			},
+		],
 	},
 	"gemini-2.5-pro-preview-05-06": {
 		maxTokens: 65_535,
 		contextWindow: 1_048_576,
 		supportsImages: true,
 		supportsPromptCache: true,
-		inputPrice: 2.5,
-		outputPrice: 15,
+		inputPrice: 1.25,
+		outputPrice: 10,
+		cacheReadsPrice: 0.125,
+		cacheWritesPrice: 1.625,
+		tiers: [
+			{
+				contextWindow: 1_000_000,
+				inputPrice: 2.5,
+				outputPrice: 15,
+				cacheReadsPrice: 0.25,
+				cacheWritesPrice: 2.875,
+			},
+		],
 	},
 	"gemini-2.5-pro-preview-06-05": {
 		maxTokens: 65_535,
 		contextWindow: 1_048_576,
 		supportsImages: true,
 		supportsPromptCache: true,
-		inputPrice: 2.5,
-		outputPrice: 15,
+		inputPrice: 1.25,
+		outputPrice: 10,
+		cacheReadsPrice: 0.125,
+		cacheWritesPrice: 1.625,
 		maxThinkingTokens: 32_768,
 		supportsReasoningBudget: true,
+		tiers: [
+			{
+				contextWindow: 1_000_000,
+				inputPrice: 2.5,
+				outputPrice: 15,
+				cacheReadsPrice: 0.25,
+				cacheWritesPrice: 2.875,
+			},
+		],
 	},
 	"gemini-2.5-pro": {
 		maxTokens: 64_000,
 		contextWindow: 1_048_576,
 		supportsImages: true,
 		supportsPromptCache: true,
-		inputPrice: 2.5,
-		outputPrice: 15,
+		inputPrice: 1.25,
+		outputPrice: 10,
+		cacheReadsPrice: 0.125,
+		cacheWritesPrice: 1.625,
 		maxThinkingTokens: 32_768,
 		supportsReasoningBudget: true,
 		requiredReasoningBudget: true,
 		tiers: [
 			{
-				contextWindow: 200_000,
-				inputPrice: 1.25,
-				outputPrice: 10,
-				cacheReadsPrice: 0.31,
-			},
-			{
-				contextWindow: Infinity,
+				contextWindow: 1_000_000,
 				inputPrice: 2.5,
 				outputPrice: 15,
-				cacheReadsPrice: 0.625,
+				cacheReadsPrice: 0.25,
+				cacheWritesPrice: 2.875,
 			},
 		],
 	},
@@ -173,6 +203,15 @@ export const vertexModels = {
 		cacheWritesPrice: 3.75,
 		cacheReadsPrice: 0.3,
 		supportsReasoningBudget: true,
+		tiers: [
+			{
+				contextWindow: 1_000_000,
+				inputPrice: 6.0,
+				outputPrice: 22.5,
+				cacheWritesPrice: 7.5,
+				cacheReadsPrice: 0.6,
+			},
+		],
 	},
 	"claude-sonnet-4-5@20250929": {
 		maxTokens: 8192,
@@ -184,6 +223,15 @@ export const vertexModels = {
 		cacheWritesPrice: 3.75,
 		cacheReadsPrice: 0.3,
 		supportsReasoningBudget: true,
+		tiers: [
+			{
+				contextWindow: 1_000_000,
+				inputPrice: 6.0,
+				outputPrice: 22.5,
+				cacheWritesPrice: 7.5,
+				cacheReadsPrice: 0.6,
+			},
+		],
 	},
 	"claude-haiku-4-5@20251001": {
 		maxTokens: 8192,
@@ -216,6 +264,15 @@ export const vertexModels = {
 		outputPrice: 75.0,
 		cacheWritesPrice: 18.75,
 		cacheReadsPrice: 1.5,
+		tiers: [
+			{
+				contextWindow: 1_000_000,
+				inputPrice: 30.0,
+				outputPrice: 112.5,
+				cacheWritesPrice: 37.5,
+				cacheReadsPrice: 3.0,
+			},
+		],
 	},
 	"claude-3-7-sonnet@20250219:thinking": {
 		maxTokens: 64_000,
diff --git a/src/api/providers/fetchers/__tests__/openrouter.spec.ts b/src/api/providers/fetchers/__tests__/openrouter.spec.ts
index 37cdc5443980..3a1393a7ee45 100644
--- a/src/api/providers/fetchers/__tests__/openrouter.spec.ts
+++ b/src/api/providers/fetchers/__tests__/openrouter.spec.ts
@@ -93,10 +93,19 @@ describe("OpenRouter API", () => {
 					inputPrice: 1.25,
 					outputPrice: 10,
 					cacheWritesPrice: 1.625,
-					cacheReadsPrice: 0.31,
+					cacheReadsPrice: 0.125,
 					description: undefined,
 					supportsReasoningEffort: undefined,
 					supportedParameters: undefined,
+					tiers: [
+						{
+							contextWindow: 1_000_000,
+							inputPrice: 2.5,
+							outputPrice: 15,
+							cacheReadsPrice: 0.25,
+							cacheWritesPrice: 2.875,
+						},
+					],
 				},
 				"google-ai-studio": {
 					maxTokens: 65536,
@@ -107,10 +116,19 @@ describe("OpenRouter API", () => {
 					inputPrice: 1.25,
 					outputPrice: 10,
 					cacheWritesPrice: 1.625,
-					cacheReadsPrice: 0.31,
+					cacheReadsPrice: 0.125,
 					description: undefined,
 					supportsReasoningEffort: undefined,
 					supportedParameters: undefined,
+					tiers: [
+						{
+							contextWindow: 1_000_000,
+							inputPrice: 2.5,
+							outputPrice: 15,
+							cacheReadsPrice: 0.25,
+							cacheWritesPrice: 2.875,
+						},
+					],
 				},
 			})
 
diff --git a/src/api/providers/fetchers/openrouter.ts b/src/api/providers/fetchers/openrouter.ts
index b546c40a3cfc..c058e747bb81 100644
--- a/src/api/providers/fetchers/openrouter.ts
+++ b/src/api/providers/fetchers/openrouter.ts
@@ -263,5 +263,56 @@ export const parseOpenRouterModel = ({
 		modelInfo.maxTokens = 32768
 	}
 
+	// Add tiered pricing for Gemini 2.5 Pro models on OpenRouter
+	if (id.includes("gemini-2.5-pro") || id.includes("gemini/2.5-pro")) {
+		modelInfo.inputPrice = 1.25
+		modelInfo.outputPrice = 10
+		modelInfo.cacheReadsPrice = 0.125
+		modelInfo.cacheWritesPrice = 1.625
+		modelInfo.tiers = [
+			{
+				contextWindow: 1_000_000,
+				inputPrice: 2.5,
+				outputPrice: 15,
+				cacheReadsPrice: 0.25,
+				cacheWritesPrice: 2.875,
+			},
+		]
+	}
+
+	// Add tiered pricing for Claude Sonnet 4 and 4.5 on OpenRouter
+	if (id === "anthropic/claude-sonnet-4" || id === "anthropic/claude-sonnet-4.5") {
+		modelInfo.inputPrice = 3.0
+		modelInfo.outputPrice = 15.0
+		modelInfo.cacheWritesPrice = 3.75
+		modelInfo.cacheReadsPrice = 0.3
+		modelInfo.tiers = [
+			{
+				contextWindow: 1_000_000,
+				inputPrice: 6.0,
+				outputPrice: 22.5,
+				cacheWritesPrice: 7.5,
+				cacheReadsPrice: 0.6,
+			},
+		]
+	}
+
+	// Add tiered pricing for Qwen 3 Max on OpenRouter
+	if (id.toLowerCase().includes("qwen") && id.toLowerCase().includes("max")) {
+		modelInfo.inputPrice = 1.2
+		modelInfo.outputPrice = 6
+		modelInfo.cacheReadsPrice = 0.24
+		modelInfo.cacheWritesPrice = 0 // Free
+		modelInfo.tiers = [
+			{
+				contextWindow: 1_000_000,
+				inputPrice: 3,
+				outputPrice: 15,
+				cacheReadsPrice: 0.6,
+				cacheWritesPrice: 0, // Free
+			},
+		]
+	}
+
 	return modelInfo
 }
diff --git a/src/shared/cost.ts b/src/shared/cost.ts
index fea686d8aed8..76aecf3754df 100644
--- a/src/shared/cost.ts
+++ b/src/shared/cost.ts
@@ -6,6 +6,65 @@ export interface ApiCostResult {
 	totalCost: number
 }
 
+/**
+ * Finds the appropriate pricing tier based on the total input tokens.
+ * Returns the prices from the matching tier, or the base prices if no tiers are defined.
+ */
+function getTieredPricing(
+	modelInfo: ModelInfo,
+	totalInputTokens: number,
+): {
+	inputPrice: number | undefined
+	outputPrice: number | undefined
+	cacheWritesPrice: number | undefined
+	cacheReadsPrice: number | undefined
+} {
+	// If there are no tiers defined, use the base prices
+	if (!modelInfo.tiers || modelInfo.tiers.length === 0) {
+		return {
+			inputPrice: modelInfo.inputPrice,
+			outputPrice: modelInfo.outputPrice,
+			cacheWritesPrice: modelInfo.cacheWritesPrice,
+			cacheReadsPrice: modelInfo.cacheReadsPrice,
+		}
+	}
+
+	// If within base context window, use base prices
+	if (totalInputTokens <= modelInfo.contextWindow) {
+		return {
+			inputPrice: modelInfo.inputPrice,
+			outputPrice: modelInfo.outputPrice,
+			cacheWritesPrice: modelInfo.cacheWritesPrice,
+			cacheReadsPrice: modelInfo.cacheReadsPrice,
+		}
+	}
+
+	// Find the appropriate tier based on the total input tokens
+	// Tiers are checked in order, and we use the first tier where the token count
+	// is less than or equal to the tier's context window
+	const tier = modelInfo.tiers.find((tier) => totalInputTokens <= tier.contextWindow)
+
+	if (tier) {
+		// Use tier prices, falling back to base prices if not defined in the tier
+		return {
+			inputPrice: tier.inputPrice ?? modelInfo.inputPrice,
+			outputPrice: tier.outputPrice ?? modelInfo.outputPrice,
+			cacheWritesPrice: tier.cacheWritesPrice ?? modelInfo.cacheWritesPrice,
+			cacheReadsPrice: tier.cacheReadsPrice ?? modelInfo.cacheReadsPrice,
+		}
+	}
+
+	// If no tier matches (all tiers have smaller context windows than the token count),
+	// use the last (highest) tier's prices
+	const lastTier = modelInfo.tiers[modelInfo.tiers.length - 1]
+	return {
+		inputPrice: lastTier.inputPrice ?? modelInfo.inputPrice,
+		outputPrice: lastTier.outputPrice ?? modelInfo.outputPrice,
+		cacheWritesPrice: lastTier.cacheWritesPrice ?? modelInfo.cacheWritesPrice,
+		cacheReadsPrice: lastTier.cacheReadsPrice ?? modelInfo.cacheReadsPrice,
+	}
+}
+
 function calculateApiCostInternal(
 	modelInfo: ModelInfo,
 	inputTokens: number,
@@ -15,10 +74,13 @@ function calculateApiCostInternal(
 	totalInputTokens: number,
 	totalOutputTokens: number,
 ): ApiCostResult {
-	const cacheWritesCost = ((modelInfo.cacheWritesPrice || 0) / 1_000_000) * cacheCreationInputTokens
-	const cacheReadsCost = ((modelInfo.cacheReadsPrice || 0) / 1_000_000) * cacheReadInputTokens
-	const baseInputCost = ((modelInfo.inputPrice || 0) / 1_000_000) * inputTokens
-	const outputCost = ((modelInfo.outputPrice || 0) / 1_000_000) * outputTokens
+	// Get the appropriate prices based on the total input tokens (for tiered pricing)
+	const { inputPrice, outputPrice, cacheWritesPrice, cacheReadsPrice } = getTieredPricing(modelInfo, totalInputTokens)
+
+	const cacheWritesCost = ((cacheWritesPrice || 0) / 1_000_000) * cacheCreationInputTokens
+	const cacheReadsCost = ((cacheReadsPrice || 0) / 1_000_000) * cacheReadInputTokens
+	const baseInputCost = ((inputPrice || 0) / 1_000_000) * inputTokens
+	const outputCost = ((outputPrice || 0) / 1_000_000) * outputTokens
 	const totalCost = cacheWritesCost + cacheReadsCost + baseInputCost + outputCost
 
 	return {