opencode-token-tracker 1.4.0 → 1.5.1

package/README.md

@@ -2,7 +2,7 @@
 
 Real-time token usage and cost tracking plugin for [OpenCode](https://opencode.ai).
 
-![Toast Screenshot](./docs/toast-demo.png)
+[English](./README.md) | [简体中文](./README.zh-CN.md)
 
 ## Features
 
@@ -14,6 +14,15 @@ Real-time token usage and cost tracking plugin for [OpenCode](https://opencode.a
 - **JSONL logging** - All usage data saved locally for analysis
 - **Multi-model support** - Claude, GPT, DeepSeek, Gemini, and more
 
+## AI Engineering Framework
+
+This project uses the [AI Engineering Framework (AIEF)](https://github.com/tongsh6/ai-engineering-framework) to organize AI collaboration context and conventions.
+
+- `AGENTS.md` defines repository-level collaboration rules
+- `context/` stores technical snapshots, coding conventions, and business semantics
+
+If you are building AI-assisted engineering workflows, we strongly recommend adopting AIEF in your own repositories for clearer context management and more consistent agent outputs.
+
 ## Installation
 
 Add to your OpenCode config file (`~/.config/opencode/opencode.json`):
@@ -291,9 +300,10 @@ All prices are in **USD per 1 million tokens**:
 Pricing is resolved in this order (first match wins):
 
 1. **Provider-level** - Override all models for a provider
-2. **User model config** - Custom model pricing in config file
-3. **Built-in pricing** - Default pricing table
-4. **Fallback** - $1/M input, $4/M output
+2. **Provider-specific model config** - Custom pricing for the same model under different providers
+3. **User model config** - Generic custom model pricing in config file
+4. **Built-in pricing** - Default pricing table
+5. **Fallback** - $1/M input, $4/M output
 
 #### Example: Free providers
 
@@ -321,6 +331,23 @@ Override or add pricing for specific models (prices in USD per 1M tokens):
 }
 ```
 
+#### Example: Same model, different provider pricing
+
+If the same model has different prices under different providers, nest provider names under the model key:
+
+```json
+{
+  "models": {
+    "deepseek/deepseek-v4-flash": {
+      "openrouter": { "input": 0.14, "output": 0.28, "cacheRead": 0.0028 },
+      "siliconflow": { "input": 0.2, "output": 0.4 }
+    }
+  }
+}
+```
+
+You can still mix this with the original flat model pricing format.
+
 ### Toast Settings
 
 | Option | Type | Default | Description |

package/README.zh-CN.md

@@ -0,0 +1,385 @@
+# opencode-token-tracker
+
+面向 [OpenCode](https://opencode.ai) 的实时 token 用量与成本追踪插件。
+
+[English](./README.md) | [简体中文](./README.zh-CN.md)
+
+## 功能特性
+
+- **实时 Toast 提示**：每次 AI 响应后展示 token 用量和成本
+- **预算控制**：支持日/周/月预算与阈值预警
+- **会话统计**：跟踪整个 session 的累计消耗
+- **CLI 统计工具**：按 day/week/month 查看统计，支持按 model/agent/provider 分组
+- **成本估算**：基于模型定价自动计算估算成本
+- **JSONL 日志**：本地持久化所有用量记录，便于分析
+- **多模型支持**：Claude、GPT、DeepSeek、Gemini 等
+
+## AI Engineering Framework
+
+本项目使用 [AI Engineering Framework (AIEF)](https://github.com/tongsh6/ai-engineering-framework) 组织 AI 协作上下文与规范。
+
+- `AGENTS.md`：仓库级 AI 协作规则
+- `context/`：技术快照、编码约定、业务语义文档
+
+如果你在构建 AI-assisted engineering 工作流，推荐在你的仓库中采用 AIEF，以获得更清晰的上下文管理与更稳定的 agent 输出。
+
+## 安装
+
+在 OpenCode 配置文件 `~/.config/opencode/opencode.json` 中添加插件：
+
+```json
+{
+  "$schema": "https://opencode.ai/config.json",
+  "plugin": ["opencode-token-tracker"]
+}
+```
+
+重启 OpenCode 后会自动安装插件。
+
+## 使用
+
+### Toast 提示
+
+安装后，每次 AI 响应会看到类似提示：
+
+```
+12.5K tokens
+$0.023 | Session: $0.156
+```
+
+配置预算后，超阈值会显示预警：
+
+```
+⚠️ Budget exceeded!
+Daily: $5.50/$5.00 (110%)
+```
+
+当会话 idle 时，会显示会话摘要：
+
+```
+Session: 45.2K tokens
+$0.156 | 8 msgs | 5min
+```
+
+### 预算控制
+
+设置预算限额，避免意外超支：
+
+```bash
+# 查看预算状态
+opencode-tokens budget
+```
+
+示例输出：
+
+```
+  Budget Status
+  ══════════════════════════════════════════════════════════════════
+
+  🟢 Daily
+    $3.50 / $10.00  [███████░░░░░░░░░░░░░] 35%
+    Remaining: $6.50
+
+  🟡 Weekly
+    $42.00 / $50.00  [████████████████░░░░] 84%
+    Remaining: $8.00
+
+  🟢 Monthly
+    $120.00 / $200.00  [████████████░░░░░░░░] 60%
+    Remaining: $80.00
+
+  Legend: 🟢 OK  🟡 Warning (>80%)  🔴 Exceeded
+```
+
+预算配置文件：`~/.config/opencode/token-tracker.json`
+
+```json
+{
+  "budget": {
+    "daily": 10,
+    "weekly": 50,
+    "monthly": 200,
+    "warnAt": 0.8
+  }
+}
+```
+
+### CLI 统计
+
+```bash
+# 全量汇总
+opencode-tokens
+
+# 今日统计
+opencode-tokens today
+
+# 本周统计（按模型分组）
+opencode-tokens week --by model
+
+# 本月统计（展示全部分组）
+opencode-tokens month --by all
+
+# 按天拆分
+opencode-tokens --by daily
+```
+
+`--by` 可选：
+- `model`：按模型分组
+- `agent`：按 agent 分组
+- `provider`：按 provider 分组
+- `daily`：按天分组
+- `all`：显示全部
+
+示例输出：
+
+```
+  Today's Usage
+  ──────────────────────────────────────────────────
+  Total Tokens:           2.81M
+    Input:                2.74M
+    Output:               72.9K
+    Reasoning:             7.1K
+  Cache Read:            12.62M
+  Total Cost:            $32.93
+  Messages:                 230
+
+  By Model
+  ─────────────────────────────────────────────────────
+  Model                Tokens        Cost    Msgs
+  ---------------  ----------  ----------  ------
+  claude-opus-4.5       2.70M      $32.93     206
+  deepseek-chat         23.4K     $0.0025       6
+  gpt-5.2               86.9K     $0.0000      18
+```
+
+### 定价与配置命令
+
+```bash
+# 查看预算状态
+opencode-tokens budget
+
+# 查看内置定价表
+opencode-tokens pricing
+
+# 查看你实际使用的模型与定价来源
+opencode-tokens models
+
+# 查看当前配置
+opencode-tokens config
+
+# 基于当前使用情况生成示例配置
+opencode-tokens config init
+```
+
+`models` 示例输出：
+
+```
+  Model                     Provider              Msgs  Pricing     
+  ------------------------  ----------------  --------  ------------
+  claude-opus-4.5           github-copilot         379  provider cfg
+  deepseek-chat             deepseek                 6  built-in    
+  gpt-5.2                   openai                  18  built-in    
+```
+
+可以帮助你了解：
+- 当前使用了哪些模型和 provider
+- 定价来源是内置表、用户配置还是默认回退
+- 需要在配置文件中补充哪些模型定价
+
+## 日志文件
+
+token 记录保存在：
+
+```
+~/.config/opencode/logs/token-tracker/tokens.jsonl
+```
+
+单行示例：
+
+```json
+{
+  "type": "tokens",
+  "sessionId": "ses_xxx",
+  "messageId": "msg_xxx",
+  "agent": "build",
+  "model": "claude-opus-4.5",
+  "provider": "github-copilot",
+  "input": 1500,
+  "output": 350,
+  "reasoning": 0,
+  "cacheRead": 5000,
+  "cacheWrite": 0,
+  "cost": 0.0234,
+  "_ts": 1234567890123
+}
+```
+
+## 支持模型
+
+| Provider | Models |
+| --- | --- |
+| Anthropic | Claude Opus 4.5, Sonnet 4/4.5, Haiku 4/4.5 |
+| OpenAI | GPT-5.x, GPT-4.x, o1, o3 |
+| DeepSeek | deepseek-chat, deepseek-reasoner |
+| Google | Gemini 2.x, 3.x |
+
+未知模型会使用默认定价估算。
+
+## 配置说明
+
+在 `~/.config/opencode/token-tracker.json` 创建配置：
+
+```json
+{
+  "providers": {
+    "github-copilot": { "input": 0, "output": 0 }
+  },
+  "models": {
+    "my-custom-model": { "input": 1, "output": 2 }
+  },
+  "toast": {
+    "enabled": true,
+    "duration": 3000,
+    "showOnIdle": true
+  },
+  "budget": {
+    "daily": 10,
+    "weekly": 50,
+    "monthly": 200,
+    "warnAt": 0.8
+  }
+}
+```
+
+### 定价字段
+
+定价单位均为 **USD / 1M tokens**：
+
+| 字段 | 说明 | 示例 |
+| --- | --- | --- |
+| `input` | 输入 token 单价 | `15` = $15 / 1M tokens |
+| `output` | 输出 token 单价 | `75` = $75 / 1M tokens |
+| `cacheRead` | 缓存读取 token 单价（可选） | `1.5` = $1.5 / 1M tokens |
+| `cacheWrite` | 缓存写入 token 单价（可选） | `18.75` = $18.75 / 1M tokens |
+
+**如何查找模型定价：**
+
+1. 查看各厂商官方定价页：
+   - [Anthropic Claude](https://www.anthropic.com/pricing)
+   - [OpenAI](https://openai.com/pricing)
+   - [DeepSeek](https://platform.deepseek.com/api-docs/pricing)
+   - [Google Gemini](https://ai.google.dev/pricing)
+
+2. 或执行 `opencode-tokens pricing` 查看内置定价表
+
+**常见场景：**
+
+| 场景 | 配置 |
+| --- | --- |
+| 订阅制服务（GitHub Copilot、Cursor） | `{ "input": 0, "output": 0 }` |
+| 免费/本地模型 | `{ "input": 0, "output": 0 }` |
+| 自定义 API（已知定价） | 查看 provider 官方定价页 |
+
+### 定价优先级
+
+定价解析顺序（命中即止）：
+
+1. **Provider 覆盖** — 为某个 provider 的所有模型统一设置
+2. **按 provider 的 model 配置** — 同一模型在不同 provider 下使用不同定价
+3. **用户 model 配置** — 为特定模型自定义通用定价
+4. **内置定价** — 默认定价表
+5. **默认回退** — $1/M input，$4/M output
+
+#### 示例：免费 provider
+
+使用 GitHub Copilot 等订阅制服务时，将成本设为 $0：
+
+```json
+{
+  "providers": {
+    "github-copilot": { "input": 0, "output": 0 },
+    "cursor": { "input": 0, "output": 0 }
+  }
+}
+```
+
+#### 示例：自定义模型定价
+
+为特定模型覆盖或新增定价（单位 USD / 1M tokens）：
+
+```json
+{
+  "models": {
+    "claude-opus-4.5": { "input": 12, "output": 60, "cacheRead": 1.2 },
+    "my-local-model": { "input": 0, "output": 0 }
+  }
+}
+```
+
+#### 示例：同一模型在不同 provider 下使用不同价格
+
+如果同一个模型在不同 provider 下价格不同，可以在模型名下继续按 provider 配置：
+
+```json
+{
+  "models": {
+    "deepseek/deepseek-v4-flash": {
+      "openrouter": { "input": 0.14, "output": 0.28, "cacheRead": 0.0028 },
+      "siliconflow": { "input": 0.2, "output": 0.4 }
+    }
+  }
+}
+```
+
+这种写法可以和原来的扁平 `models` 配置同时使用。
+
+### 预算设置
+
+| 选项 | 类型 | 默认值 | 说明 |
+| --- | --- | --- | --- |
+| `daily` | number | - | 每日预算上限（USD） |
+| `weekly` | number | - | 每周预算上限（USD） |
+| `monthly` | number | - | 每月预算上限（USD） |
+| `warnAt` | number | `0.8` | 预警阈值（0-1），如 0.8 = 达到 80% 时预警 |
+
+超出预算时：
+- Toast 提示切换为预警/错误样式
+- 使用 `opencode-tokens budget` 查看详细状态
+- 预算在每日零点（daily）、每周一（weekly）、每月 1 日（monthly）自动重置
+
+### Toast 设置
+
+| 选项 | 类型 | 默认值 | 说明 |
+| --- | --- | --- | --- |
+| `enabled` | boolean | `true` | 是否显示 Toast 提示 |
+| `duration` | number | `3000` | Toast 显示时长（毫秒） |
+| `showOnIdle` | boolean | `true` | 会话 idle 时是否显示会话摘要 |
+
+## 开发
+
+```bash
+# 克隆仓库
+git clone https://github.com/tongsh6/opencode-token-tracker.git
+cd opencode-token-tracker
+
+# 安装依赖
+npm install
+
+# 构建
+npm run build
+
+# 本地联调
+npm link
+cd ~/.config/opencode
+npm link opencode-token-tracker
+```
+
+## License
+
+MIT © [tongsh6](https://github.com/tongsh6)
+
+## Related
+
+- [OpenCode](https://opencode.ai) - AI coding assistant
+- [OpenCode Plugins](https://opencode.ai/docs/plugins) - 插件文档
+- [oh-my-opencode](https://github.com/code-yeongyu/oh-my-opencode) - OpenCode 增强插件

package/dist/bin/opencode-tokens.js

@@ -1,95 +1,20 @@
 #!/usr/bin/env node
+import { BUILTIN_PRICING, DEFAULT_CONFIG, findModelConfigPricing, formatCost, formatTokens, getStartOfDay, getStartOfWeek, getStartOfMonth, validateConfig } from "../lib/shared.js";
 import { readFileSync, existsSync, writeFileSync } from "fs";
 import { join } from "path";
 import { homedir } from "os";
 const CONFIG_DIR = join(homedir(), ".config", "opencode");
 const CONFIG_FILE = join(CONFIG_DIR, "token-tracker.json");
 const LOG_FILE = join(CONFIG_DIR, "logs", "token-tracker", "tokens.jsonl");
-// Built-in pricing (USD per 1M tokens) - Updated 2026-02-05
-// Keep in sync with index.ts BUILTIN_PRICING
-const BUILTIN_PRICING = {
-    // Anthropic Claude (https://www.anthropic.com/pricing#api)
-    "claude-opus-4.5": { input: 5, output: 25, cacheRead: 0.5, cacheWrite: 6.25 },
-    "claude-sonnet-4.5": { input: 3, output: 15, cacheRead: 0.3, cacheWrite: 3.75 },
-    "claude-sonnet-4": { input: 3, output: 15, cacheRead: 0.3, cacheWrite: 3.75 },
-    "claude-haiku-4.5": { input: 1, output: 5, cacheRead: 0.1, cacheWrite: 1.25 },
-    "claude-haiku-4": { input: 1, output: 5, cacheRead: 0.1, cacheWrite: 1.25 },
-    "claude-opus-4.1": { input: 15, output: 75, cacheRead: 1.5, cacheWrite: 18.75 },
-    "claude-opus-4": { input: 15, output: 75, cacheRead: 1.5, cacheWrite: 18.75 },
-    "claude-haiku-3": { input: 0.25, output: 1.25, cacheRead: 0.03, cacheWrite: 0.3 },
-    // OpenAI GPT (https://openai.com/api/pricing/)
-    "gpt-5.2": { input: 1.75, output: 14, cacheRead: 0.175 },
-    "gpt-5.2-pro": { input: 21, output: 168 },
-    "gpt-5-mini": { input: 0.25, output: 2, cacheRead: 0.025 },
-    "gpt-5.1": { input: 2, output: 8 },
-    "gpt-5": { input: 5, output: 15 },
-    "gpt-4.1": { input: 3, output: 12, cacheRead: 0.75 },
-    "gpt-4.1-mini": { input: 0.8, output: 3.2, cacheRead: 0.2 },
-    "gpt-4.1-nano": { input: 0.2, output: 0.8, cacheRead: 0.05 },
-    "gpt-4o": { input: 2.5, output: 10 },
-    "gpt-4o-mini": { input: 0.15, output: 0.6 },
-    "o3": { input: 10, output: 40 },
-    "o3-mini": { input: 1.1, output: 4.4 },
-    "o4-mini": { input: 4, output: 16, cacheRead: 1 },
-    "o1": { input: 15, output: 60 },
-    "o1-mini": { input: 1.1, output: 4.4 },
-    // DeepSeek (https://api-docs.deepseek.com/quick_start/pricing)
-    "deepseek-chat": { input: 0.28, output: 0.42, cacheRead: 0.028 },
-    "deepseek-reasoner": { input: 0.28, output: 0.42, cacheRead: 0.028 },
-    // Google Gemini (https://cloud.google.com/vertex-ai/generative-ai/pricing)
-    "gemini-3-pro": { input: 2, output: 12, cacheRead: 0.2 },
-    "gemini-3-pro-preview": { input: 2, output: 12, cacheRead: 0.2 },
-    "gemini-3-flash": { input: 0.5, output: 2, cacheRead: 0.05 },
-    "gemini-3-flash-preview": { input: 0.5, output: 2, cacheRead: 0.05 },
-    "gemini-2.5-pro": { input: 1.25, output: 10, cacheRead: 0.125 },
-    "gemini-2.5-flash": { input: 0.1, output: 0.4, cacheRead: 0.01 },
-    "gemini-2.5-flash-lite": { input: 0.1, output: 0.4, cacheRead: 0.01 },
-    "gemini-2.0-flash": { input: 0.15, output: 0.6, cacheRead: 0.015 },
-    // Fallback
-    "_default": { input: 1, output: 4 },
-};
 // ============================================================================
 // Helpers
 // ============================================================================
-function formatTokens(tokens) {
-    if (tokens >= 1_000_000)
-        return `${(tokens / 1_000_000).toFixed(2)}M`;
-    if (tokens >= 1_000)
-        return `${(tokens / 1_000).toFixed(1)}K`;
-    return tokens.toString();
-}
-function formatCost(cost) {
-    if (cost < 0.01)
-        return `$${cost.toFixed(4)}`;
-    if (cost < 1)
-        return `$${cost.toFixed(3)}`;
-    return `$${cost.toFixed(2)}`;
-}
 function padRight(str, len) {
     return str.length >= len ? str : str + " ".repeat(len - str.length);
 }
 function padLeft(str, len) {
     return str.length >= len ? str : " ".repeat(len - str.length) + str;
 }
-function getStartOfDay(date) {
-    const d = new Date(date);
-    d.setHours(0, 0, 0, 0);
-    return d.getTime();
-}
-function getStartOfWeek(date) {
-    const d = new Date(date);
-    const day = d.getDay();
-    const diff = d.getDate() - day + (day === 0 ? -6 : 1);
-    d.setDate(diff);
-    d.setHours(0, 0, 0, 0);
-    return d.getTime();
-}
-function getStartOfMonth(date) {
-    const d = new Date(date);
-    d.setDate(1);
-    d.setHours(0, 0, 0, 0);
-    return d.getTime();
-}
 // ============================================================================
 // Data Loading
 // ============================================================================
@@ -120,11 +45,20 @@ function loadEntries(since) {
 function loadConfig() {
     try {
         if (existsSync(CONFIG_FILE)) {
-            return JSON.parse(readFileSync(CONFIG_FILE, "utf-8"));
+            const raw = JSON.parse(readFileSync(CONFIG_FILE, "utf-8"));
+            const result = validateConfig(raw);
+            if (result.warnings.length > 0) {
+                for (const w of result.warnings) {
+                    console.error(`  [token-tracker] config warning: ${w}`);
+                }
+            }
+            return result.config;
         }
     }
-    catch { }
-    return {};
+    catch {
+        console.error("  [token-tracker] config warning: Config file is not valid JSON, using defaults");
+    }
+    return DEFAULT_CONFIG;
 }
 // ============================================================================
 // Stats Aggregation
@@ -171,13 +105,13 @@ function printSummary(title, stats) {
     console.log();
     console.log(`  ${title}`);
     console.log(`  ${"─".repeat(50)}`);
-    console.log(`  Total Tokens:    ${padLeft(formatTokens(totalTokens), 12)}`);
-    console.log(`    Input:         ${padLeft(formatTokens(stats.input), 12)}`);
-    console.log(`    Output:        ${padLeft(formatTokens(stats.output), 12)}`);
+    console.log(`  Total Tokens:    ${padLeft(formatTokens(totalTokens, 2), 12)}`);
+    console.log(`    Input:         ${padLeft(formatTokens(stats.input, 2), 12)}`);
+    console.log(`    Output:        ${padLeft(formatTokens(stats.output, 2), 12)}`);
     if (stats.reasoning > 0) {
-        console.log(`    Reasoning:     ${padLeft(formatTokens(stats.reasoning), 12)}`);
+        console.log(`    Reasoning:     ${padLeft(formatTokens(stats.reasoning, 2), 12)}`);
     }
-    console.log(`  Cache Read:      ${padLeft(formatTokens(stats.cacheRead), 12)}`);
+    console.log(`  Cache Read:      ${padLeft(formatTokens(stats.cacheRead, 2), 12)}`);
     console.log(`  Total Cost:      ${padLeft(formatCost(stats.cost), 12)}`);
     console.log(`  Messages:        ${padLeft(stats.count.toString(), 12)}`);
     console.log();
@@ -199,7 +133,7 @@ function printTable(title, groups, labelHeader) {
     console.log(`  ${"-".repeat(labelWidth)}  ${"-".repeat(tokensWidth)}  ${"-".repeat(costWidth)}  ${"-".repeat(countWidth)}`);
     for (const [label, stats] of sorted) {
         const totalTokens = stats.input + stats.output;
-        console.log(`  ${padRight(label, labelWidth)}  ${padLeft(formatTokens(totalTokens), tokensWidth)}  ${padLeft(formatCost(stats.cost), costWidth)}  ${padLeft(stats.count.toString(), countWidth)}`);
+        console.log(`  ${padRight(label, labelWidth)}  ${padLeft(formatTokens(totalTokens, 2), tokensWidth)}  ${padLeft(formatCost(stats.cost), costWidth)}  ${padLeft(stats.count.toString(), countWidth)}`);
     }
     console.log();
 }
@@ -224,7 +158,7 @@ function printDailyBreakdown(entries) {
     console.log(`  ${"-".repeat(dateWidth)}  ${"-".repeat(tokensWidth)}  ${"-".repeat(costWidth)}  ${"-".repeat(countWidth)}`);
     for (const [date, stats] of sorted) {
         const totalTokens = stats.input + stats.output;
-        console.log(`  ${padRight(date, dateWidth)}  ${padLeft(formatTokens(totalTokens), tokensWidth)}  ${padLeft(formatCost(stats.cost), costWidth)}  ${padLeft(stats.count.toString(), countWidth)}`);
+        console.log(`  ${padRight(date, dateWidth)}  ${padLeft(formatTokens(totalTokens, 2), tokensWidth)}  ${padLeft(formatCost(stats.cost), costWidth)}  ${padLeft(stats.count.toString(), countWidth)}`);
     }
     console.log();
 }
@@ -285,15 +219,15 @@ function cmdStats(period, breakdown) {
 function cmdPricing() {
     const config = loadConfig();
     console.log(`
-  Built-in Pricing Table (USD per 1M tokens) - Updated 2026-02-05
+  Built-in Pricing Table (USD per 1M tokens) - Updated 2026-02-11
   ══════════════════════════════════════════════════════════════════
 `);
     // Group by provider
     const groups = {
-        "Anthropic Claude": ["claude-opus-4.5", "claude-sonnet-4.5", "claude-sonnet-4", "claude-haiku-4.5", "claude-haiku-4", "claude-opus-4.1", "claude-opus-4", "claude-haiku-3"],
+        "Anthropic Claude": ["claude-opus-4.6", "claude-opus-4.5", "claude-sonnet-4.5", "claude-sonnet-4", "claude-haiku-4.5", "claude-haiku-4", "claude-opus-4.1", "claude-opus-4", "claude-haiku-3"],
         "OpenAI": ["gpt-5.2", "gpt-5.2-pro", "gpt-5-mini", "gpt-5.1", "gpt-5", "gpt-4.1", "gpt-4.1-mini", "gpt-4.1-nano", "gpt-4o", "gpt-4o-mini", "o3", "o3-mini", "o4-mini", "o1", "o1-mini"],
         "DeepSeek": ["deepseek-chat", "deepseek-reasoner"],
-        "Google Gemini": ["gemini-3-pro", "gemini-3-pro-preview", "gemini-3-flash", "gemini-3-flash-preview", "gemini-2.5-pro", "gemini-2.5-flash", "gemini-2.5-flash-lite", "gemini-2.0-flash"],
+        "Google Gemini": ["gemini-3-pro", "gemini-3-pro-preview", "gemini-3-flash", "gemini-3-flash-preview", "gemini-2.5-pro", "gemini-2.5-flash", "gemini-2.5-flash-lite", "gemini-2.0-flash", "gemini-2.0-flash-lite"],
     };
     const modelWidth = 20;
     const priceWidth = 10;
@@ -358,7 +292,7 @@ function cmdModels() {
         if (config.providers?.[provider]) {
             status = "provider cfg";
         }
-        else if (config.models?.[model]) {
+        else if (findModelConfigPricing(config.models, model, provider)) {
             status = "model cfg";
         }
         else if (!BUILTIN_PRICING[model]) {
@@ -548,10 +482,14 @@ function cmdBudget() {
             return "🟡";
         return "🟢";
     };
-    const entries = loadEntries();
+    // Calculate the earliest period start to minimize data loaded
+    const dayStart = getStartOfDay(now);
+    const weekStart = getStartOfWeek(now);
+    const monthStart = getStartOfMonth(now);
+    const earliestSince = Math.min(budget.daily ? dayStart : Infinity, budget.weekly ? weekStart : Infinity, budget.monthly ? monthStart : Infinity);
+    const entries = loadEntries(earliestSince);
     // Daily budget
     if (budget.daily) {
-        const dayStart = getStartOfDay(now);
         const dayEntries = entries.filter(e => e._ts >= dayStart);
         const spent = dayEntries.reduce((sum, e) => sum + (e.cost ?? 0), 0);
         const pct = spent / budget.daily;
@@ -563,7 +501,6 @@ function cmdBudget() {
     }
     // Weekly budget
     if (budget.weekly) {
-        const weekStart = getStartOfWeek(now);
         const weekEntries = entries.filter(e => e._ts >= weekStart);
         const spent = weekEntries.reduce((sum, e) => sum + (e.cost ?? 0), 0);
         const pct = spent / budget.weekly;
@@ -575,7 +512,6 @@ function cmdBudget() {
     }
     // Monthly budget
     if (budget.monthly) {
-        const monthStart = getStartOfMonth(now);
         const monthEntries = entries.filter(e => e._ts >= monthStart);
         const spent = monthEntries.reduce((sum, e) => sum + (e.cost ?? 0), 0);
         const pct = spent / budget.monthly;