Merge branch 'main' into snyk-upgrade-aws-sdk-packages

Author: RomneyDa

.github/workflows/jetbrains-release.yaml

@@ -491,7 +491,7 @@ jobs:
 
       # Run Qodana inspections
       - name: Qodana - Code Inspection
-        uses: JetBrains/qodana-action@v2025.2.2
+        uses: JetBrains/qodana-action@v2025.2.3
         with:
           cache-default-branch-only: true
 

.github/workflows/runloop-blueprint-template.json

@@ -1,8 +1,24 @@
 {
   "name": "cn",
+  "dockerfile": "FROM runloop:runloop/universal-ubuntu-24.04-x86_64-dnd",
   "system_setup_commands": [
     "npm i -g @continuedev/cli@latest",
     "sudo apt update",
-    "sudo apt install -y ripgrep"
-  ]
+    "printf '#!/bin/sh\\nexit 101\\n' | sudo tee /usr/sbin/policy-rc.d > /dev/null && sudo chmod +x /usr/sbin/policy-rc.d",
+    "sudo apt install -y --no-install-recommends ripgrep chromium chromium-driver xvfb",
+    "sudo rm /usr/sbin/policy-rc.d",
+    "sudo mkdir -p /opt/google/chrome",
+    "sudo ln -s /usr/bin/chromium /opt/google/chrome/chrome",
+    "curl -1sLf 'https://artifacts-cli.infisical.com/setup.deb.sh' | sudo -E bash",
+    "sudo apt-get install -y infisical"
+  ],
+  "launch_parameters": {
+    "launch_commands": [
+      "export DISPLAY=:99",
+      "nohup Xvfb :99 -screen 0 1920x1080x24 > /tmp/xvfb.log 2>&1 &",
+      "sleep 2",
+      "sudo mkdir -p /home/user",
+      "for dir in /root/*; do [ -d \"$dir\" ] && [ -d \"$dir/.git\" ] && sudo ln -sf \"$dir\" \"/home/user/$(basename \"$dir\")\"; done"
+    ]
+  }
 }

.github/workflows/upload-runloop-blueprint.yml

@@ -0,0 +1,59 @@
+name: Upload Runloop Blueprint
+
+on:
+  workflow_dispatch:
+  push:
+    branches:
+      - main
+    paths:
+      - '.github/workflows/runloop-blueprint-template.json'
+
+permissions:
+  contents: read
+
+jobs:
+  upload-blueprint:
+    name: Upload Blueprint to Runloop
+    runs-on: ubuntu-latest
+    # Only run this workflow on the main repository (continuedev/continue)
+    if: github.repository == 'continuedev/continue'
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v6
+
+      - name: Publish Blueprint to Runloop
+        env:
+          RUNLOOP_API_KEY: ${{ secrets.RUNLOOP_API_KEY }}
+        run: |
+          echo "Publishing blueprint 'cn' to Runloop API"
+
+          # Use blueprint configuration from template file
+          cp .github/workflows/runloop-blueprint-template.json blueprint.json
+
+          # Publish to Runloop API
+          response=$(curl -X POST "https://api.runloop.ai/v1/blueprints" \
+            -H "Authorization: Bearer $RUNLOOP_API_KEY" \
+            -H "Content-Type: application/json" \
+            -d @blueprint.json \
+            -w "%{http_code}" \
+            -s)
+
+          http_code=$(echo "$response" | tail -c 4)
+          response_body=$(echo "$response" | head -c -4)
+
+          if [[ "$http_code" == "200" ]] || [[ "$http_code" == "201" ]]; then
+            echo "✅ Successfully published blueprint to Runloop API"
+            echo "Response: $response_body"
+
+            # Extract blueprint ID if available
+            blueprint_id=$(echo "$response_body" | jq -r '.id // empty')
+            if [[ -n "$blueprint_id" ]]; then
+              echo "Blueprint ID: $blueprint_id"
+              echo "blueprint_id=$blueprint_id" >> $GITHUB_OUTPUT
+            fi
+          else
+            echo "❌ Failed to publish blueprint to Runloop API"
+            echo "HTTP Code: $http_code"
+            echo "Response: $response_body"
+            exit 1
+          fi

core/index.d.ts

@@ -267,6 +267,11 @@ export interface IContextProvider {
   get deprecationMessage(): string | null;
 }
 
+export interface SessionUsage extends Usage {
+  /** Total cumulative cost in USD for all LLM API calls in this session */
+  totalCost: number;
+}
+
 export interface Session {
   sessionId: string;
   title: string;
@@ -276,6 +281,8 @@ export interface Session {
   mode?: MessageModes;
   /** Optional: title of the selected chat model for this session */
   chatModelTitle?: string | null;
+  /** Optional: cumulative usage and cost for all LLM API calls in this session */
+  usage?: SessionUsage;
 }
 
 export interface BaseSessionMetadata {

core/llm/utils/calculateRequestCost.ts

@@ -12,19 +12,21 @@ function calculateAnthropicCost(
   // Normalize model name to handle various formats
   const normalizedModel = model.toLowerCase();
 
-  // Define pricing per million tokens (MTok)
+  // Define pricing per million tokens (MTok) by model family prefix
   const pricing: Record<
     string,
     { input: number; output: number; cacheWrite: number; cacheRead: number }
   > = {
-    // Claude Opus 4 (most intelligent model)
-    "claude-3-opus": {
-      input: 15,
-      output: 75,
-      cacheWrite: 18.75,
-      cacheRead: 1.5,
+    // Claude Opus 4.5 (most intelligent model)
+    "claude-opus-4-5": {
+      input: 5,
+      output: 25,
+      cacheWrite: 6.25,
+      cacheRead: 0.5,
     },
-    "claude-3-opus-20240229": {
+
+    // Claude Opus 4 (legacy)
+    "claude-3-opus": {
       input: 15,
       output: 75,
       cacheWrite: 18.75,
@@ -38,18 +40,6 @@ function calculateAnthropicCost(
       cacheWrite: 3.75,
       cacheRead: 0.3,
     },
-    "claude-3-5-sonnet-20241022": {
-      input: 3,
-      output: 15,
-      cacheWrite: 3.75,
-      cacheRead: 0.3,
-    },
-    "claude-3-5-sonnet-20240620": {
-      input: 3,
-      output: 15,
-      cacheWrite: 3.75,
-      cacheRead: 0.3,
-    },
 
     // Claude Haiku 3.5 (fastest, most cost-effective)
     "claude-3-5-haiku": {
@@ -58,12 +48,6 @@ function calculateAnthropicCost(
       cacheWrite: 1,
       cacheRead: 0.08,
     },
-    "claude-3-5-haiku-20241022": {
-      input: 0.8,
-      output: 4,
-      cacheWrite: 1,
-      cacheRead: 0.08,
-    },
 
     // Legacy Claude 3 Haiku
     "claude-3-haiku": {
@@ -72,35 +56,16 @@ function calculateAnthropicCost(
       cacheWrite: 0.3,
       cacheRead: 0.03,
     },
-    "claude-3-haiku-20240307": {
-      input: 0.25,
-      output: 1.25,
-      cacheWrite: 0.3,
-      cacheRead: 0.03,
-    },
   };
 
-  // Find matching pricing model
-  let modelPricing = pricing[normalizedModel];
+  // Sort keys by length (longest first) to match most specific patterns first
+  const sortedKeys = Object.keys(pricing).sort((a, b) => b.length - a.length);
 
-  // If exact match not found, try to find by partial match
-  if (!modelPricing) {
-    for (const [key, value] of Object.entries(pricing)) {
-      if (normalizedModel.includes(key) || key.includes(normalizedModel)) {
-        modelPricing = value;
-        break;
-      }
-    }
-  }
-
-  // If still no match, try common patterns
-  if (!modelPricing) {
-    if (normalizedModel.includes("opus")) {
-      modelPricing = pricing["claude-3-opus"];
-    } else if (normalizedModel.includes("sonnet")) {
-      modelPricing = pricing["claude-3-5-sonnet"];
-    } else if (normalizedModel.includes("haiku")) {
-      modelPricing = pricing["claude-3-5-haiku"];
+  let modelPricing = null;
+  for (const prefix of sortedKeys) {
+    if (normalizedModel.startsWith(prefix)) {
+      modelPricing = pricing[prefix];
+      break;
     }
   }
 
@@ -167,14 +132,90 @@ function calculateAnthropicCost(
   };
 }
 
+function calculateOpenAICost(
+  model: string,
+  usage: Usage,
+): CostBreakdown | null {
+  // Normalize model name
+  const normalizedModel = model.toLowerCase();
+
+  // Define pricing per million tokens (MTok) by model family prefix
+  const pricing: Record<string, { input: number; output: number }> = {
+    // GPT-4o models (most specific first)
+    "gpt-4o-mini": { input: 0.15, output: 0.6 },
+    "gpt-4o": { input: 2.5, output: 10 },
+
+    // GPT-4 Turbo models
+    "gpt-4-turbo": { input: 10, output: 30 },
+
+    // GPT-3.5 Turbo models (most specific first)
+    "gpt-3.5-turbo-0125": { input: 0.5, output: 1.5 },
+    "gpt-3.5-turbo-1106": { input: 1, output: 2 },
+    "gpt-3.5-turbo": { input: 1.5, output: 2 },
+
+    // Base GPT-4 (fallback for other gpt-4 variants)
+    "gpt-4": { input: 30, output: 60 },
+  };
+
+  // Sort keys by length (longest first) to match most specific patterns first
+  const sortedKeys = Object.keys(pricing).sort((a, b) => b.length - a.length);
+
+  let modelPricing = null;
+  for (const prefix of sortedKeys) {
+    if (normalizedModel.startsWith(prefix)) {
+      modelPricing = pricing[prefix];
+      break;
+    }
+  }
+
+  if (!modelPricing) {
+    return null; // Unknown model
+  }
+
+  // Calculate costs
+  const inputCost = (usage.promptTokens / 1_000_000) * modelPricing.input;
+  const outputCost = (usage.completionTokens / 1_000_000) * modelPricing.output;
+
+  // Build breakdown components
+  const breakdownParts: string[] = [];
+
+  if (usage.promptTokens > 0) {
+    breakdownParts.push(
+      `Input: ${usage.promptTokens.toLocaleString()} tokens × $${modelPricing.input}/MTok = $${inputCost.toFixed(6)}`,
+    );
+  }
+
+  if (usage.completionTokens > 0) {
+    breakdownParts.push(
+      `Output: ${usage.completionTokens.toLocaleString()} tokens × $${modelPricing.output}/MTok = $${outputCost.toFixed(6)}`,
+    );
+  }
+
+  const totalCost = inputCost + outputCost;
+
+  // Build final breakdown string
+  let breakdown = `Model: ${model}\n`;
+  breakdown += breakdownParts.join("\n");
+  if (breakdownParts.length > 1) {
+    breakdown += `\nTotal: $${totalCost.toFixed(6)}`;
+  }
+
+  return {
+    cost: totalCost,
+    breakdown,
+  };
+}
+
 export function calculateRequestCost(
   provider: string,
   model: string,
   usage: Usage,
 ): CostBreakdown | null {
-  switch (provider) {
+  switch (provider.toLowerCase()) {
     case "anthropic":
       return calculateAnthropicCost(model, usage);
+    case "openai":
+      return calculateOpenAICost(model, usage);
     default:
       return null;
   }