improvement(usage): update usage limit in realtime, standardize token output object across providers (#2553)

* improvement(usage-limit): update usage in real time, fix token output object

* updated tokenBreakdown to tokens, standardized input/output/total token object type across providers

* update remaining references

* ack PR comment

* remove singleton query client instance from hooks, leave only in zustand

apps/sim/lib/logs/execution/logger.ts

@@ -59,8 +59,12 @@ export class ExecutionLogger implements IExecutionLoggerService {
           output: (merged[model].output || 0) + (costs.output || 0),
           total: (merged[model].total || 0) + (costs.total || 0),
           tokens: {
-            prompt: (merged[model].tokens?.prompt || 0) + (costs.tokens?.prompt || 0),
-            completion: (merged[model].tokens?.completion || 0) + (costs.tokens?.completion || 0),
+            input:
+              (merged[model].tokens?.input || merged[model].tokens?.prompt || 0) +
+              (costs.tokens?.input || costs.tokens?.prompt || 0),
+            output:
+              (merged[model].tokens?.output || merged[model].tokens?.completion || 0) +
+              (costs.tokens?.output || costs.tokens?.completion || 0),
             total: (merged[model].tokens?.total || 0) + (costs.tokens?.total || 0),
           },
         }
@@ -195,7 +199,7 @@ export class ExecutionLogger implements IExecutionLoggerService {
           input: number
           output: number
           total: number
-          tokens: { prompt: number; completion: number; total: number }
+          tokens: { input: number; output: number; total: number }
         }
       >
     }
@@ -269,8 +273,12 @@ export class ExecutionLogger implements IExecutionLoggerService {
           input: (existingCost.input || 0) + costSummary.totalInputCost,
           output: (existingCost.output || 0) + costSummary.totalOutputCost,
           tokens: {
-            prompt: (existingCost.tokens?.prompt || 0) + costSummary.totalPromptTokens,
-            completion: (existingCost.tokens?.completion || 0) + costSummary.totalCompletionTokens,
+            input:
+              (existingCost.tokens?.input || existingCost.tokens?.prompt || 0) +
+              costSummary.totalPromptTokens,
+            output:
+              (existingCost.tokens?.output || existingCost.tokens?.completion || 0) +
+              costSummary.totalCompletionTokens,
             total: (existingCost.tokens?.total || 0) + costSummary.totalTokens,
           },
           models: this.mergeCostModels(existingCost.models || {}, costSummary.models),
@@ -280,8 +288,8 @@ export class ExecutionLogger implements IExecutionLoggerService {
           input: costSummary.totalInputCost,
           output: costSummary.totalOutputCost,
           tokens: {
-            prompt: costSummary.totalPromptTokens,
-            completion: costSummary.totalCompletionTokens,
+            input: costSummary.totalPromptTokens,
+            output: costSummary.totalCompletionTokens,
             total: costSummary.totalTokens,
           },
           models: costSummary.models,
@@ -307,9 +315,9 @@ export class ExecutionLogger implements IExecutionLoggerService {
         executionData: {
           traceSpans: redactedTraceSpans,
           finalOutput: redactedFinalOutput,
-          tokenBreakdown: {
-            prompt: mergedCost.tokens.prompt,
-            completion: mergedCost.tokens.completion,
+          tokens: {
+            input: mergedCost.tokens.input,
+            output: mergedCost.tokens.output,
             total: mergedCost.tokens.total,
           },
           models: mergedCost.models,
@@ -508,7 +516,7 @@ export class ExecutionLogger implements IExecutionLoggerService {
           input: number
           output: number
           total: number
-          tokens: { prompt: number; completion: number; total: number }
+          tokens: { input: number; output: number; total: number }
         }
       >
     },

apps/sim/lib/logs/execution/logging-factory.ts

@@ -83,7 +83,7 @@ export function calculateCostSummary(traceSpans: any[]): {
       input: number
       output: number
       total: number
-      tokens: { prompt: number; completion: number; total: number }
+      tokens: { input: number; output: number; total: number }
     }
   >
 } {
@@ -131,7 +131,7 @@ export function calculateCostSummary(traceSpans: any[]): {
       input: number
       output: number
       total: number
-      tokens: { prompt: number; completion: number; total: number }
+      tokens: { input: number; output: number; total: number }
     }
   > = {}
 
@@ -150,14 +150,14 @@ export function calculateCostSummary(traceSpans: any[]): {
           input: 0,
           output: 0,
           total: 0,
-          tokens: { prompt: 0, completion: 0, total: 0 },
+          tokens: { input: 0, output: 0, total: 0 },
         }
       }
       models[model].input += span.cost.input || 0
       models[model].output += span.cost.output || 0
       models[model].total += span.cost.total || 0
-      models[model].tokens.prompt += span.tokens?.input ?? span.tokens?.prompt ?? 0
-      models[model].tokens.completion += span.tokens?.output ?? span.tokens?.completion ?? 0
+      models[model].tokens.input += span.tokens?.input ?? span.tokens?.prompt ?? 0
+      models[model].tokens.output += span.tokens?.output ?? span.tokens?.completion ?? 0
       models[model].tokens.total += span.tokens?.total || 0
     }
   }

apps/sim/lib/logs/execution/trace-spans/trace-spans.test.ts

@@ -23,7 +23,7 @@ describe('buildTraceSpans', () => {
           output: {
             content: 'Agent response',
             model: 'gpt-4o',
-            tokens: { prompt: 10, completion: 20, total: 30 },
+            tokens: { input: 10, output: 20, total: 30 },
             providerTiming: {
               duration: 8000,
               startTime: '2024-01-01T10:00:00.000Z',
@@ -138,7 +138,7 @@ describe('buildTraceSpans', () => {
           output: {
             content: 'Agent response',
             model: 'gpt-4o',
-            tokens: { prompt: 10, completion: 20, total: 30 },
+            tokens: { input: 10, output: 20, total: 30 },
             providerTiming: {
               duration: 4000,
               startTime: '2024-01-01T10:00:00.500Z',
@@ -427,8 +427,8 @@ describe('buildTraceSpans', () => {
           output: {
             content: 'Based on my research using multiple sources...',
             model: 'gpt-4o',
-            tokens: { prompt: 50, completion: 200, total: 250 },
-            cost: { total: 0.0025, prompt: 0.001, completion: 0.0015 },
+            tokens: { input: 50, output: 200, total: 250 },
+            cost: { total: 0.0025, input: 0.001, output: 0.0015 },
             providerTiming: {
               duration: 15000,
               startTime: '2024-01-01T10:00:00.000Z',

apps/sim/lib/logs/types.ts

@@ -15,8 +15,8 @@ export interface PricingInfo {
 }
 
 export interface TokenUsage {
-  prompt: number
-  completion: number
+  input: number
+  output: number
   total: number
 }
 
@@ -102,6 +102,17 @@ export interface WorkflowExecutionLog {
     environment?: ExecutionEnvironment
     trigger?: ExecutionTrigger
     traceSpans?: TraceSpan[]
+    tokens?: { input?: number; output?: number; total?: number }
+    models?: Record<
+      string,
+      {
+        input?: number
+        output?: number
+        total?: number
+        tokens?: { input?: number; output?: number; total?: number }
+      }
+    >
+    finalOutput?: any
     errorDetails?: {
       blockId: string
       blockName: string
@@ -114,14 +125,14 @@ export interface WorkflowExecutionLog {
     input?: number
     output?: number
     total?: number
-    tokens?: { prompt?: number; completion?: number; total?: number }
+    tokens?: { input?: number; output?: number; total?: number }
     models?: Record<
       string,
       {
         input?: number
         output?: number
         total?: number
-        tokens?: { prompt?: number; completion?: number; total?: number }
+        tokens?: { input?: number; output?: number; total?: number }
       }
     >
   }