chore: resolve merge conflicts from main

- Resolved conflicts in calculateApiCostOpenAI signature (added reasoningTokens param)
- Updated cerebras.ts to use new cost calculation with reasoning tokens
- Updated groq.ts to use new cost calculation with reasoning tokens
- Updated lite-llm.ts to use new cost calculation with reasoning tokens
- Updated openai-native.ts to use new cost calculation with reasoning tokens
- All providers now properly handle reasoning token costs

Author: hannesrudolph

src/api/providers/cerebras.ts

@@ -331,7 +331,7 @@ export class CerebrasHandler extends BaseProvider implements SingleCompletionHan
 		const { info } = this.getModel()
 		// Use actual token usage from the last request
 		const { inputTokens, outputTokens } = this.lastUsage
-		const { totalCost } = calculateApiCostOpenAI(info, inputTokens, outputTokens)
+		const totalCost = calculateApiCostOpenAI(info, inputTokens, outputTokens)
 		return totalCost
 	}
 }

src/api/providers/groq.ts

@@ -64,7 +64,7 @@ export class GroqHandler extends BaseOpenAiCompatibleProvider<GroqModelId> {
 		const cacheWriteTokens = 0
 
 		// Calculate cost using OpenAI-compatible cost calculation
-		const { totalCost } = calculateApiCostOpenAI(info, inputTokens, outputTokens, cacheWriteTokens, cacheReadTokens)
+		const totalCost = calculateApiCostOpenAI(info, inputTokens, outputTokens, cacheWriteTokens, cacheReadTokens)
 
 		yield {
 			type: "usage",

src/api/providers/lite-llm.ts

@@ -165,7 +165,7 @@ export class LiteLLMHandler extends RouterProvider implements SingleCompletionHa
 					(lastUsage as any).prompt_cache_hit_tokens ||
 					0
 
-				const { totalCost } = calculateApiCostOpenAI(
+				const totalCost = calculateApiCostOpenAI(
 					info,
 					lastUsage.prompt_tokens || 0,
 					lastUsage.completion_tokens || 0,

src/api/providers/openai-native.ts

@@ -100,7 +100,7 @@ export class OpenAiNativeHandler extends BaseProvider implements SingleCompletio
 
 		// Pass total input tokens directly to calculateApiCostOpenAI
 		// The function handles subtracting both cache reads and writes internally
-		const { totalCost } = calculateApiCostOpenAI(
+		const totalCost = calculateApiCostOpenAI(
 			effectiveInfo,
 			totalInputTokens,
 			totalOutputTokens,

src/shared/cost.ts

@@ -12,20 +12,15 @@ function calculateApiCostInternal(
 	outputTokens: number,
 	cacheCreationInputTokens: number,
 	cacheReadInputTokens: number,
-	totalInputTokens: number,
-	totalOutputTokens: number,
-): ApiCostResult {
+	totalInputTokens: number, // kept for potential future use
+	totalOutputTokens: number, // kept for potential future use
+): number {
 	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
 	const totalCost = cacheWritesCost + cacheReadsCost + baseInputCost + outputCost
-
-	return {
-		totalInputTokens,
-		totalOutputTokens,
-		totalCost,
-	}
+	return totalCost
 }
 
 // For Anthropic compliant usage, the input tokens count does NOT include the
@@ -36,7 +31,7 @@ export function calculateApiCostAnthropic(
 	outputTokens: number,
 	cacheCreationInputTokens?: number,
 	cacheReadInputTokens?: number,
-): ApiCostResult {
+): number {
 	const cacheCreation = cacheCreationInputTokens || 0
 	const cacheRead = cacheReadInputTokens || 0
 
@@ -62,7 +57,7 @@ export function calculateApiCostOpenAI(
 	outputTokens: number,
 	cacheCreationInputTokens?: number,
 	cacheReadInputTokens?: number,
-): ApiCostResult {
+): number {
 	const cacheCreationInputTokensNum = cacheCreationInputTokens || 0
 	const cacheReadInputTokensNum = cacheReadInputTokens || 0
 	const nonCachedInputTokens = Math.max(0, inputTokens - cacheCreationInputTokensNum - cacheReadInputTokensNum)