@jeffreycao/copilot-api 1.14.9 → 1.14.10

package/README.md

@@ -162,7 +162,7 @@ To get started, run the `start` command with the `--claude-code` flag:
 npx @jeffreycao/copilot-api@latest start --claude-code
 ```
 
-You will be prompted to select a primary model and a "small, fast" model for background tasks. After selecting the models, a command will be copied to your clipboard. This command sets the necessary environment variables for Claude Code to use the gateway.
+You will no longer be prompted to pick models manually. The gateway automatically detects the latest available model for each Claude Code size tier — opus maps to the newest Opus model, sonnet to the newest Sonnet model, and haiku to the newest Haiku model — and generates a command that sets `ANTHROPIC_DEFAULT_OPUS_MODEL`, `ANTHROPIC_DEFAULT_SONNET_MODEL`, and `ANTHROPIC_DEFAULT_HAIKU_MODEL` accordingly. Any tier without a matching model available is omitted. The command is copied to your clipboard and sets the environment variables needed for Claude Code to use the gateway.
 
 Paste and run this command in a new terminal to launch Claude Code.
 
@@ -177,9 +177,12 @@ Here is an example `.claude/settings.json` file:
   "env": {
     "ANTHROPIC_BASE_URL": "http://localhost:4141",
     "ANTHROPIC_AUTH_TOKEN": "dummy",
-    "ANTHROPIC_MODEL": "deepseek/deepseek-v4-pro",
-    "ANTHROPIC_DEFAULT_SONNET_MODEL": "deepseek/deepseek-v4-pro",
-    "ANTHROPIC_DEFAULT_HAIKU_MODEL": "deepseek/deepseek-v4-flash",
+    "ANTHROPIC_MODEL": "gpt-5.6-sol[1m]",
+    "ANTHROPIC_DEFAULT_SONNET_MODEL": "gpt-5.6-sol[1m]",
+    "ANTHROPIC_DEFAULT_HAIKU_MODEL": "gpt-5.6-luna[1m]",
+    "CLAUDE_CODE_AUTO_COMPACT_WINDOW": "272000",
+    "CLAUDE_CODE_USE_VERTEX": "0",
+    "CLAUDE_CODE_USE_BEDROCK": "0",
     "DISABLE_NON_ESSENTIAL_MODEL_CALLS": "1",
     "CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC": "1",
     "CLAUDE_CODE_ATTRIBUTION_HEADER": "0",
@@ -187,15 +190,21 @@ Here is an example `.claude/settings.json` file:
     "CLAUDE_CODE_DISABLE_TERMINAL_TITLE": "true",
     "CLAUDE_CODE_ENABLE_AWAY_SUMMARY": "0"
   },
-  "permissions": {
-    "deny": [
-      "mcp__ide__executeCode"
-    ]
-  }
+  "alwaysThinkingEnabled": true,
+  "effortLevel": "xhigh",
+  "showThinkingSummaries": true
 }
 ```
 
 - Replace `ANTHROPIC_MODEL`, `ANTHROPIC_DEFAULT_OPUS_MODEL`, `ANTHROPIC_DEFAULT_SONNET_MODEL`, and `ANTHROPIC_DEFAULT_HAIKU_MODEL` according to your needs. After configuration, please install the claude code plugin [Plugin Integrations](#plugin-integrations).  
+- If you are using the codex provider, it is recommended **not** to configure the model name in the `codex/xxx` format (e.g. `codex/gpt-5.6-sol`). Claude Code treats the `codex/` prefix as a special pattern and applies degraded behavior — for example, it strips all previously returned thinking blocks on every request. Use the plain model name (e.g. `gpt-5.6-sol`) instead, and add a `modelMappings` entry in `config.json` to route it back to the codex provider:
+  ```json
+  "modelMappings": {
+    "gpt-5.6-sol": "codex/gpt-5.6-sol",
+    "gpt-5.6-terra": "codex/gpt-5.6-terra",
+    "gpt-5.6-luna": "codex/gpt-5.6-luna"
+  },
+  ```
 - Setting CLAUDE_CODE_ATTRIBUTION_HEADER to 0 can prevent Claude code from adding billing and version information in system prompts, thereby avoiding prompt cache invalidation.
 - Turning off CLAUDE_CODE_ENABLE_PROMPT_SUGGESTION and CLAUDE_CODE_ENABLE_AWAY_SUMMARY can prevent quota from being consumed unnecessarily.
 - Claude Code WebSearch is supported for pure search requests. For Copilot, keep the global `messageApiWebSearchModel` set to a Responses-capable GPT model or a `provider/model` alias. For provider routes, use a native Anthropic provider or an `openai-responses` provider. Add `WebSearch` to `permissions.deny` only if you want to forbid this traffic.
@@ -310,7 +319,8 @@ enabled = false
 ```
 
 > [!NOTE]
-> This configuration is specific to Codex and the GitHub Copilot provider. `name` must be set to `"OpenAI"`. It can help mitigate Codex local compact cache miss issues. 
+> This configuration is specific to Codex and the GitHub Copilot provider. `name` must be set to `"OpenAI"`. It can help mitigate Codex local compact cache miss issues.
+> If you are using the codex provider, it is recommended to set `base_url` to `"http://localhost:4141/codex"`.
 
 ## GPT Tool Search
 
@@ -610,7 +620,7 @@ Use `copilot-api auth login --provider custom` to add or update another third-pa
     }
   }
   ```
-  Built-in token prices cover Codex GPT models in USD, DashScope `qwen3.7-max`, `qwen3.7-plus`, `glm-5.1`, `glm-5.2` in CNY, DeepSeek `deepseek-v4-flash`, `deepseek-v4-pro`, `deepseek-chat`, `deepseek-reasoner` in CNY, and OpenCode Go models (`glm-5.2`, `deepseek-v4-flash`, `deepseek-v4-pro`, `kimi-k2.7-code`, `mimo-v2.5`, `mimo-v2.5-pro`, `qwen3.7-plus`, `qwen3.7-max`, `minimax-m2.5`, `minimax-m3`) in USD. User `pricing` entries override built-ins. For DashScope, cached tokens are charged as explicit cache reads when the upstream usage includes `cache_creation_input_tokens`; otherwise `cachedInput` is used as the implicit cache read price. For DeepSeek, `prompt_cache_hit_tokens` map to cached input and `prompt_cache_miss_tokens` map to regular input.
+  Built-in token prices cover Codex GPT models in USD, DashScope `qwen3.7-max`, `qwen3.7-plus`, `glm-5.1`, `glm-5.2` in CNY, DeepSeek `deepseek-v4-flash`, `deepseek-v4-pro`, `deepseek-chat`, `deepseek-reasoner` in CNY, and OpenCode Go models (`glm-5.2`, `grok-4.5`, `deepseek-v4-flash`, `deepseek-v4-pro`, `kimi-k2.7-code`, `kimi-k3`, `mimo-v2.5`, `mimo-v2.5-pro`, `qwen3.7-plus`, `qwen3.7-max`, `minimax-m2.5`, `minimax-m3`) in USD. User `pricing` entries override built-ins. For DashScope, cached tokens are charged as explicit cache reads when the upstream usage includes `cache_creation_input_tokens`; otherwise `cachedInput` is used as the implicit cache read price. For DeepSeek, `prompt_cache_hit_tokens` map to cached input and `prompt_cache_miss_tokens` map to regular input.
 - **smallModel:** Fallback model used for tool-less warmup messages (e.g., Claude Code probe requests); defaults to gpt-5-mini.
 - **contextManagement:** Controls whether the proxy adds Responses API `context_management` compaction instructions. `messages` applies when Anthropic-style `/v1/messages` requests are translated to Responses API, including `openai-responses` provider message routes, and defaults to `true`. `responses` applies to native `/v1/responses` traffic, including `provider/model` aliases and the built-in `codex` provider, and defaults to `false`. Enable `responses` only after checking that your client supports context management compaction. When enabled, the request includes `context_management` in the body and keeps only the latest compaction carrier on follow-up turns. **Note:** Context management is forcibly disabled for GPT-5.6 and above models (e.g. `gpt-5.6-sol`, `gpt-5.6-terra`, `gpt-5.6-luna`) because enabling it breaks prompt cache hits on those models. This override takes precedence over the `contextManagement` and `modelResponsesApiCompactThresholds` settings.
 - **modelResponsesApiCompactThresholds:** Per-model Responses API `compact_threshold` overrides used when the proxy adds `context_management`. These values take precedence over the fallback threshold from `resolveResponsesCompactThreshold` (`max_prompt_tokens * ratio`, or the default fallback). Defaults set `gpt-5.4` and `gpt-5.5` to `217600` (`272000 * 0.8`), and `gpt-5.6-sol`, `gpt-5.6-terra`, `gpt-5.6-luna` to `231200` (`272000 * 0.85`). Models not listed continue to use the normal fallback logic.
@@ -619,7 +629,7 @@ Use `copilot-api auth login --provider custom` to add or update another third-pa
   - **Forwarding:** the resolved value remains `output_config.effort` for the Copilot native Messages API and becomes `reasoning.effort` when translated to the Responses API.
   - **Configuration values:** `none`, `minimal`, `low`, `medium`, `high`, `xhigh`, and `max`.
 - **useMessagesApi:** When `true`, Claude-family models that support Copilot's native `/v1/messages` endpoint will use the Messages API; otherwise they fall back to `/chat/completions`. Set to `false` to disable Messages API routing and always use `/chat/completions`. Defaults to `true`.
-- **useResponsesApiWebSocket:** When `true`, Responses API requests use Copilot's websocket transport for models that advertise `ws:/responses`; models that only advertise `/responses` continue to use HTTP. Set to `false` to disable websocket routing and use HTTP `/responses` whenever the selected model supports it. Defaults to `true`.
+- **useResponsesApiWebSocket:** When `true`, Responses API requests use Copilot's websocket transport for models that advertise `ws:/responses`; models that only advertise `/responses` continue to use HTTP. Set to `false` to disable websocket routing and use HTTP `/responses` whenever the selected model supports it. Defaults to `true`. If the Responses API WebSocket gets closed, it is usually caused by your own network. If you are using a VPN, try switching to a different node.
 - **useResponsesApiWebSearch:** When `true`, the server keeps Responses API tools with `type: "web_search"` and forwards them upstream. Set to `false` to strip those tools from `/responses` payloads. Defaults to `true`.
 - **messageApiWebSearchModel:** Global fallback model used when a top-level Copilot `/v1/messages` request contains only the server-side `web_search` tool. Defaults to `gpt-5-mini`. If the value is a `provider/model` alias, the request is routed into that provider's Messages API path with the provider prefix stripped. For Copilot GPT models, web search runs through `/responses`. Mixed `web_search` plus custom tools are not supported and the server-side `web_search` tool is stripped.
 - **claudeTokenMultiplier:** Multiplier applied to the fallback GPT-tokenizer estimate for Claude `/v1/messages/count_tokens` requests. Defaults to `1.15`. Increase it if your client is still compacting too late. This setting is only used when the proxy is estimating Claude tokens locally; if `anthropicApiKey` is configured and Anthropic token counting succeeds, the exact Anthropic count is returned instead.

package/README.zh-CN.md

@@ -162,7 +162,7 @@ https://github.com/caozhiyuan/copilot-api/releases
 npx @jeffreycao/copilot-api@latest start --claude-code
 ```
 
-你会被提示选择一个主模型，以及一个用于后台任务的 "small, fast" 模型。选择完成后，会有一条命令被复制到剪贴板中。该命令会设置 Claude Code 使用这个 AI gateway 所需的环境变量。
+你不再需要手动选择模型。Gateway 会自动检测每个 Claude Code 尺寸档位对应的最新可用模型——opus 映射到最新的 Opus 模型，sonnet 映射到最新的 Sonnet 模型，haiku 映射到最新的 Haiku 模型——并生成相应设置 `ANTHROPIC_DEFAULT_OPUS_MODEL`、`ANTHROPIC_DEFAULT_SONNET_MODEL` 和 `ANTHROPIC_DEFAULT_HAIKU_MODEL` 的命令。若某个档位没有匹配的可用模型，则会被省略。该命令会被复制到剪贴板，并设置 Claude Code 使用这个 AI gateway 所需的环境变量。
 
 在新的终端中粘贴并执行这条命令，即可启动 Claude Code。
 
@@ -179,9 +179,12 @@ npx @jeffreycao/copilot-api@latest start --claude-code
   "env": {
     "ANTHROPIC_BASE_URL": "http://localhost:4141",
     "ANTHROPIC_AUTH_TOKEN": "dummy",
-    "ANTHROPIC_MODEL": "deepseek/deepseek-v4-pro",
-    "ANTHROPIC_DEFAULT_SONNET_MODEL": "deepseek/deepseek-v4-pro",
-    "ANTHROPIC_DEFAULT_HAIKU_MODEL": "deepseek/deepseek-v4-flash",
+    "ANTHROPIC_MODEL": "gpt-5.6-sol[1m]",
+    "ANTHROPIC_DEFAULT_SONNET_MODEL": "gpt-5.6-sol[1m]",
+    "ANTHROPIC_DEFAULT_HAIKU_MODEL": "gpt-5.6-luna[1m]",
+    "CLAUDE_CODE_AUTO_COMPACT_WINDOW": "272000",
+    "CLAUDE_CODE_USE_VERTEX": "0",
+    "CLAUDE_CODE_USE_BEDROCK": "0",
     "DISABLE_NON_ESSENTIAL_MODEL_CALLS": "1",
     "CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC": "1",
     "CLAUDE_CODE_ATTRIBUTION_HEADER": "0",
@@ -189,15 +192,21 @@ npx @jeffreycao/copilot-api@latest start --claude-code
     "CLAUDE_CODE_DISABLE_TERMINAL_TITLE": "true",
     "CLAUDE_CODE_ENABLE_AWAY_SUMMARY": "0"
   },
-  "permissions": {
-    "deny": [
-      "mcp__ide__executeCode"
-    ]
-  }
+  "alwaysThinkingEnabled": true,
+  "effortLevel": "xhigh",
+  "showThinkingSummaries": true
 }
 ```
 
 - 请根据需要替换 `ANTHROPIC_MODEL`、`ANTHROPIC_DEFAULT_OPUS_MODEL`、`ANTHROPIC_DEFAULT_SONNET_MODEL` 和 `ANTHROPIC_DEFAULT_HAIKU_MODEL`。配置完成后，请安装 claude code 插件，见 [插件集成](#plugin-integrations)。
+- 如果你使用的是 codex provider，建议**不要**将模型名配置成 `codex/xxx` 格式（如 `codex/gpt-5.6-sol`）。Claude Code 会针对 `codex/` 前缀做降智行为——例如每次请求时移除所有之前返回的思考块（thinking blocks）。请使用纯模型名（如 `gpt-5.6-sol`），并在 `config.json` 中配置 `modelMappings` 将其映射回 codex provider：
+  ```json
+  "modelMappings": {
+    "gpt-5.6-sol": "codex/gpt-5.6-sol",
+    "gpt-5.6-terra": "codex/gpt-5.6-terra",
+    "gpt-5.6-luna": "codex/gpt-5.6-luna"
+  },
+  ```
 - 将 `CLAUDE_CODE_ATTRIBUTION_HEADER` 设为 `0` 可以阻止 Claude Code 在 system prompt 中附加计费和版本信息，从而避免 prompt cache 失效。
 - 关闭 `CLAUDE_CODE_ENABLE_PROMPT_SUGGESTION` 和 `CLAUDE_CODE_ENABLE_AWAY_SUMMARY` 可以避免不必要地消耗额度。
 - Claude Code WebSearch 已支持纯搜索请求。Copilot 路径请保持全局 `messageApiWebSearchModel` 指向 Responses-capable GPT 模型或 `provider/model` 别名；provider 路由请使用原生 Anthropic provider 或 `openai-responses` provider。只有在你明确想禁止这类流量时，才需要把 `WebSearch` 加到 `permissions.deny`。
@@ -313,6 +322,7 @@ enabled = false
 
 > [!NOTE]
 > 此配置仅限于 Codex 与 GitHub Copilot provider。`name` 一定要配置为 `"OpenAI"`。它可以缓解 Codex local compact 不命中缓存的问题。
+> 如果使用 codex provider，建议将 `base_url` 配置为 `"http://localhost:4141/codex"`。
 
 ## GPT Tool Search
 
@@ -616,7 +626,7 @@ Copilot API 现在使用子命令结构，主要命令包括：
     }
   }
   ```
-  内置 token 价格覆盖 Codex GPT 模型（USD）、DashScope `qwen3.7-max`、`qwen3.7-plus`、`glm-5.1`、`glm-5.2`（CNY），DeepSeek `deepseek-v4-flash`、`deepseek-v4-pro`、`deepseek-chat`、`deepseek-reasoner`（CNY），以及 OpenCode Go 模型（`glm-5.2`、`deepseek-v4-flash`、`deepseek-v4-pro`、`kimi-k2.7-code`、`mimo-v2.5`、`mimo-v2.5-pro`、`qwen3.7-plus`、`qwen3.7-max`、`minimax-m2.5`、`minimax-m3`，USD）。用户配置的 `pricing` 优先于内置价格。DashScope 若上游 usage 中出现 `cache_creation_input_tokens` 字段，cached tokens 按显式缓存读价计费；否则 `cachedInput` 作为隐式缓存读价。DeepSeek 的 `prompt_cache_hit_tokens` 会归入 cached input，`prompt_cache_miss_tokens` 会归入普通 input。
+  内置 token 价格覆盖 Codex GPT 模型（USD）、DashScope `qwen3.7-max`、`qwen3.7-plus`、`glm-5.1`、`glm-5.2`（CNY），DeepSeek `deepseek-v4-flash`、`deepseek-v4-pro`、`deepseek-chat`、`deepseek-reasoner`（CNY），以及 OpenCode Go 模型（`glm-5.2`、`grok-4.5`、`deepseek-v4-flash`、`deepseek-v4-pro`、`kimi-k2.7-code`、`kimi-k3`、`mimo-v2.5`、`mimo-v2.5-pro`、`qwen3.7-plus`、`qwen3.7-max`、`minimax-m2.5`、`minimax-m3`，USD）。用户配置的 `pricing` 优先于内置价格。DashScope 若上游 usage 中出现 `cache_creation_input_tokens` 字段，cached tokens 按显式缓存读价计费；否则 `cachedInput` 作为隐式缓存读价。DeepSeek 的 `prompt_cache_hit_tokens` 会归入 cached input，`prompt_cache_miss_tokens` 会归入普通 input。
 - **smallModel：** 无工具预热消息的回退模型（例如 Claude Code 的探测请求）；默认是 `gpt-5-mini`。
 - **contextManagement：** 控制代理是否为 Responses API 附加 `context_management` 压缩指令。`messages` 作用于被翻译成 Responses API 的 Anthropic 风格 `/v1/messages` 请求，包括 `openai-responses` provider 的 Messages 路由，默认值为 `true`。`responses` 作用于 native `/v1/responses` 流量，包括 `provider/model` 别名和内置 `codex` provider，默认值为 `false`。只有在确认客户端支持 context management compaction 后，才建议在 Responses API 下启用 `responses`。启用后，请求体会带上 `context_management`，并在后续轮次中仅保留最新的压缩承载内容。**注意：** 对于 GPT-5.6 及以上模型（如 `gpt-5.6-sol`、`gpt-5.6-terra`、`gpt-5.6-luna`），context management 功能会被强制禁用，因为开启后会破坏这些模型的 prompt 缓存命中。此强制覆盖优先于 `contextManagement` 和 `modelResponsesApiCompactThresholds` 配置。
 - **modelResponsesApiCompactThresholds：** 按模型覆盖 Responses API 的 `compact_threshold`，仅在代理自动附加 `context_management` 时使用。它的优先级高于 `resolveResponsesCompactThreshold` 基于 `max_prompt_tokens * ratio` 的兜底阈值。默认将 `gpt-5.4` 和 `gpt-5.5` 设为 `217600`（`272000 * 0.8`），将 `gpt-5.6-sol`、`gpt-5.6-terra`、`gpt-5.6-luna` 设为 `231200`（`272000 * 0.85`）。未列出的模型继续使用原有兜底逻辑。
@@ -625,7 +635,7 @@ Copilot API 现在使用子命令结构，主要命令包括：
   - **转发字段：** 走 Copilot 原生 Messages API 时，最终值写入 `output_config.effort`；转换为 Responses API 时，最终值写入 `reasoning.effort`。
   - **配置可选值：** `none`、`minimal`、`low`、`medium`、`high`、`xhigh`、`max`。
 - **useMessagesApi：** 当为 `true` 时，支持 Copilot 原生 `/v1/messages` 的 Claude 系模型会走 Messages API；否则回退到 `/chat/completions`。设为 `false` 可禁用 Messages API 路由，始终使用 `/chat/completions`。默认值为 `true`。
-- **useResponsesApiWebSocket：** 当为 `true` 时，Responses API 请求会优先对声明了 `ws:/responses` 的模型使用 Copilot websocket transport；仅声明 `/responses` 的模型仍走 HTTP。设为 `false` 可禁用 websocket 路由，并在模型支持 `/responses` 时使用 HTTP `/responses`。默认值为 `true`。
+- **useResponsesApiWebSocket：** 当为 `true` 时，Responses API 请求会优先对声明了 `ws:/responses` 的模型使用 Copilot websocket transport；仅声明 `/responses` 的模型仍走 HTTP。设为 `false` 可禁用 websocket 路由，并在模型支持 `/responses` 时使用 HTTP `/responses`。默认值为 `true`。如果遇到 Responses API WebSocket closed，一般是自己的网络问题。如果使用了 VPN，建议切换节点。
 - **useResponsesApiWebSearch：** 当为 `true` 时，服务端会保留 Responses API 中 `type: "web_search"` 的工具并透传到上游。设为 `false` 则会从 `/responses` payload 中移除这些工具。默认值为 `true`。
 - **messageApiWebSearchModel：** 顶层 Copilot `/v1/messages` 请求只包含服务端 `web_search` 工具时使用的全局模型，默认值为 `gpt-5-mini`。如果该值是 `provider/model` 别名，请求会进入对应 provider 的 Messages API 路径，并在转发前移除 provider 前缀。对于 Copilot GPT 模型，web search 会通过 `/responses` 执行。混合 `web_search` 与自定义工具的场景暂不支持，服务端会移除 server-side `web_search`。
 - **claudeTokenMultiplier：** 用于 Claude `/v1/messages/count_tokens` 请求在本地走 GPT tokenizer 估算时的乘数。默认值为 `1.15`。如果你的客户端仍然过晚触发上下文压缩，可以适当调大。这个配置只会在代理本地估算 Claude token 时生效；如果已经配置 `anthropicApiKey` 且 Anthropic token counting 调用成功，则会直接返回 Anthropic 的精确计数，不会使用这个乘数。

package/dist/main.js

@@ -25,7 +25,7 @@ bindElectronFetch();
 const { auth } = await import("./auth-DvJ8r_tN.js");
 const { debug } = await import("./debug-Bb5hamXX.js");
 const { mcp } = await import("./mcp-BseuqgHR.js");
-const { start } = await import("./start-B0sr92AZ.js");
+const { start } = await import("./start-BBonobw6.js");
 await runMain(defineCommand({
 	meta: {
 		name: "copilot-api",

package/dist/models-haiuFowF.js

@@ -0,0 +1,90 @@
+import { z as state } from "./token-DHWoZfxf.js";
+//#region src/lib/models.ts
+/**
+* Converts a Copilot upstream model ID to a client-friendly ID that Claude Code
+* and Claude Desktop recognize (dots in version replaced with hyphens).
+* e.g. "claude-sonnet-4.6" -> "claude-sonnet-4-6"
+* Non-Claude models are returned unchanged.
+*/
+const toClientModelId = (modelId) => {
+	const normalized = normalizeSdkModelId(modelId);
+	if (!normalized) return modelId;
+	const versionHyphenated = normalized.version.replaceAll(".", "-");
+	return `claude-${normalized.family}-${versionHyphenated}`;
+};
+const findEndpointModel = (sdkModelId) => {
+	const models = state.models?.data ?? [];
+	const exactMatch = models.find((m) => m.id === sdkModelId);
+	if (exactMatch) return exactMatch;
+	const normalized = normalizeSdkModelId(sdkModelId);
+	if (!normalized) return;
+	const modelName = `claude-${normalized.family}-${normalized.version}`;
+	const model = models.find((m) => m.id === modelName);
+	if (model) return model;
+};
+/**
+* Finds the latest available model for a given Claude family (e.g. "opus",
+* "sonnet", "haiku") among the models currently cached in `state.models`.
+* "Latest" is determined by the highest semantic version parsed from the model
+* ID. Returns `undefined` when no model of that family is available.
+*/
+const getLatestModelForFamily = (family) => {
+	const models = state.models?.data ?? [];
+	let best;
+	for (const model of models) {
+		const normalized = normalizeSdkModelId(model.id);
+		if (!normalized || normalized.family !== family) continue;
+		const [majorPart, minorPart = "0"] = normalized.version.split(".");
+		const major = Number.parseInt(majorPart, 10);
+		const minor = Number.parseInt(minorPart, 10);
+		if (Number.isNaN(major) || Number.isNaN(minor)) continue;
+		if (!best || major > best.major || major === best.major && minor > best.minor) best = {
+			model,
+			major,
+			minor
+		};
+	}
+	return best?.model;
+};
+/**
+* Normalizes an SDK model ID to extract the model family and version.
+* this method from github copilot extension
+* Examples:
+* - "claude-opus-4-5-20251101" -> { family: "opus", version: "4.5" }
+* - "claude-3-5-sonnet-20241022" -> { family: "sonnet", version: "3.5" }
+* - "claude-sonnet-4-20250514" -> { family: "sonnet", version: "4" }
+* - "claude-haiku-3-5-20250514" -> { family: "haiku", version: "3.5" }
+* - "claude-haiku-4.5" -> { family: "haiku", version: "4.5" }
+*/
+const normalizeSdkModelId = (sdkModelId) => {
+	const withoutDate = sdkModelId.toLowerCase().replace(/-\d{8}$/, "");
+	const pattern1 = withoutDate.match(/^claude-(\w+)-(\d+)\.(\d+)$/);
+	if (pattern1) return {
+		family: pattern1[1],
+		version: `${pattern1[2]}.${pattern1[3]}`
+	};
+	const pattern2 = withoutDate.match(/^claude-(\w+)-(\d+)-(\d+)$/);
+	if (pattern2) return {
+		family: pattern2[1],
+		version: `${pattern2[2]}.${pattern2[3]}`
+	};
+	const pattern3 = withoutDate.match(/^claude-(\d+)-(\d+)-(\w+)$/);
+	if (pattern3) return {
+		family: pattern3[3],
+		version: `${pattern3[1]}.${pattern3[2]}`
+	};
+	const pattern4 = withoutDate.match(/^claude-(\w+)-(\d+)$/);
+	if (pattern4) return {
+		family: pattern4[1],
+		version: pattern4[2]
+	};
+	const pattern5 = withoutDate.match(/^claude-(\d+)-(\w+)$/);
+	if (pattern5) return {
+		family: pattern5[2],
+		version: pattern5[1]
+	};
+};
+//#endregion
+export { toClientModelId as i, getLatestModelForFamily as n, normalizeSdkModelId as r, findEndpointModel as t };
+
+//# sourceMappingURL=models-haiuFowF.js.map

package/dist/models-haiuFowF.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"models-haiuFowF.js","names":[],"sources":["../src/lib/models.ts"],"sourcesContent":["import type { Model } from \"~/services/copilot/get-models\"\n\nimport { state } from \"~/lib/state\"\n\n/**\n * Converts a Copilot upstream model ID to a client-friendly ID that Claude Code\n * and Claude Desktop recognize (dots in version replaced with hyphens).\n * e.g. \"claude-sonnet-4.6\" -> \"claude-sonnet-4-6\"\n * Non-Claude models are returned unchanged.\n */\nexport const toClientModelId = (modelId: string): string => {\n  const normalized = normalizeSdkModelId(modelId)\n  if (!normalized) return modelId\n  const versionHyphenated = normalized.version.replaceAll(\".\", \"-\")\n  return `claude-${normalized.family}-${versionHyphenated}`\n}\n\nexport interface NormalizedSdkModelId {\n  family: string\n  version: string\n}\n\nexport const findEndpointModel = (sdkModelId: string): Model | undefined => {\n  const models = state.models?.data ?? []\n  const exactMatch = models.find((m) => m.id === sdkModelId)\n  if (exactMatch) {\n    return exactMatch\n  }\n\n  const normalized = normalizeSdkModelId(sdkModelId)\n  if (!normalized) {\n    return undefined\n  }\n\n  const modelName = `claude-${normalized.family}-${normalized.version}`\n  const model = models.find((m) => m.id === modelName)\n  if (model) {\n    return model\n  }\n\n  return undefined\n}\n\n/**\n * Finds the latest available model for a given Claude family (e.g. \"opus\",\n * \"sonnet\", \"haiku\") among the models currently cached in `state.models`.\n * \"Latest\" is determined by the highest semantic version parsed from the model\n * ID. Returns `undefined` when no model of that family is available.\n */\nexport const getLatestModelForFamily = (family: string): Model | undefined => {\n  const models = state.models?.data ?? []\n\n  let best: { model: Model; major: number; minor: number } | undefined\n\n  for (const model of models) {\n    const normalized = normalizeSdkModelId(model.id)\n    if (!normalized || normalized.family !== family) {\n      continue\n    }\n\n    const [majorPart, minorPart = \"0\"] = normalized.version.split(\".\")\n    const major = Number.parseInt(majorPart, 10)\n    const minor = Number.parseInt(minorPart, 10)\n    if (Number.isNaN(major) || Number.isNaN(minor)) {\n      continue\n    }\n\n    if (\n      !best\n      || major > best.major\n      || (major === best.major && minor > best.minor)\n    ) {\n      best = { model, major, minor }\n    }\n  }\n\n  return best?.model\n}\n\n/**\n * Normalizes an SDK model ID to extract the model family and version.\n * this method from github copilot extension\n * Examples:\n * - \"claude-opus-4-5-20251101\" -> { family: \"opus\", version: \"4.5\" }\n * - \"claude-3-5-sonnet-20241022\" -> { family: \"sonnet\", version: \"3.5\" }\n * - \"claude-sonnet-4-20250514\" -> { family: \"sonnet\", version: \"4\" }\n * - \"claude-haiku-3-5-20250514\" -> { family: \"haiku\", version: \"3.5\" }\n * - \"claude-haiku-4.5\" -> { family: \"haiku\", version: \"4.5\" }\n */\nexport const normalizeSdkModelId = (\n  sdkModelId: string,\n): NormalizedSdkModelId | undefined => {\n  const lower = sdkModelId.toLowerCase()\n\n  // Strip date suffix (8 digits at the end)\n  const withoutDate = lower.replace(/-\\d{8}$/, \"\")\n\n  // Pattern 1: claude-{family}-{major}.{minor} (e.g., claude-haiku-4.5)\n  const pattern1 = withoutDate.match(/^claude-(\\w+)-(\\d+)\\.(\\d+)$/)\n  if (pattern1) {\n    return { family: pattern1[1], version: `${pattern1[2]}.${pattern1[3]}` }\n  }\n\n  // Pattern 2: claude-{family}-{major}-{minor} (e.g., claude-opus-4-5, claude-haiku-3-5)\n  const pattern2 = withoutDate.match(/^claude-(\\w+)-(\\d+)-(\\d+)$/)\n  if (pattern2) {\n    return { family: pattern2[1], version: `${pattern2[2]}.${pattern2[3]}` }\n  }\n\n  // Pattern 3: claude-{major}-{minor}-{family} (e.g., claude-3-5-sonnet)\n  const pattern3 = withoutDate.match(/^claude-(\\d+)-(\\d+)-(\\w+)$/)\n  if (pattern3) {\n    return { family: pattern3[3], version: `${pattern3[1]}.${pattern3[2]}` }\n  }\n\n  // Pattern 4: claude-{family}-{major} (e.g., claude-sonnet-4)\n  const pattern4 = withoutDate.match(/^claude-(\\w+)-(\\d+)$/)\n  if (pattern4) {\n    return { family: pattern4[1], version: pattern4[2] }\n  }\n\n  // Pattern 5: claude-{major}-{family} (e.g., claude-3-opus)\n  const pattern5 = withoutDate.match(/^claude-(\\d+)-(\\w+)$/)\n  if (pattern5) {\n    return { family: pattern5[2], version: pattern5[1] }\n  }\n\n  return undefined\n}\n"],"mappings":";;;;;;;;AAUA,MAAa,mBAAmB,YAA4B;CAC1D,MAAM,aAAa,oBAAoB,QAAQ;CAC/C,IAAI,CAAC,YAAY,OAAO;CACxB,MAAM,oBAAoB,WAAW,QAAQ,WAAW,KAAK,IAAI;CACjE,OAAO,UAAU,WAAW,OAAO,GAAG;;AAQxC,MAAa,qBAAqB,eAA0C;CAC1E,MAAM,SAAS,MAAM,QAAQ,QAAQ,EAAE;CACvC,MAAM,aAAa,OAAO,MAAM,MAAM,EAAE,OAAO,WAAW;CAC1D,IAAI,YACF,OAAO;CAGT,MAAM,aAAa,oBAAoB,WAAW;CAClD,IAAI,CAAC,YACH;CAGF,MAAM,YAAY,UAAU,WAAW,OAAO,GAAG,WAAW;CAC5D,MAAM,QAAQ,OAAO,MAAM,MAAM,EAAE,OAAO,UAAU;CACpD,IAAI,OACF,OAAO;;;;;;;;AAYX,MAAa,2BAA2B,WAAsC;CAC5E,MAAM,SAAS,MAAM,QAAQ,QAAQ,EAAE;CAEvC,IAAI;CAEJ,KAAK,MAAM,SAAS,QAAQ;EAC1B,MAAM,aAAa,oBAAoB,MAAM,GAAG;EAChD,IAAI,CAAC,cAAc,WAAW,WAAW,QACvC;EAGF,MAAM,CAAC,WAAW,YAAY,OAAO,WAAW,QAAQ,MAAM,IAAI;EAClE,MAAM,QAAQ,OAAO,SAAS,WAAW,GAAG;EAC5C,MAAM,QAAQ,OAAO,SAAS,WAAW,GAAG;EAC5C,IAAI,OAAO,MAAM,MAAM,IAAI,OAAO,MAAM,MAAM,EAC5C;EAGF,IACE,CAAC,QACE,QAAQ,KAAK,SACZ,UAAU,KAAK,SAAS,QAAQ,KAAK,OAEzC,OAAO;GAAE;GAAO;GAAO;GAAO;;CAIlC,OAAO,MAAM;;;;;;;;;;;;AAaf,MAAa,uBACX,eACqC;CAIrC,MAAM,cAHQ,WAAW,aAGA,CAAC,QAAQ,WAAW,GAAG;CAGhD,MAAM,WAAW,YAAY,MAAM,8BAA8B;CACjE,IAAI,UACF,OAAO;EAAE,QAAQ,SAAS;EAAI,SAAS,GAAG,SAAS,GAAG,GAAG,SAAS;EAAM;CAI1E,MAAM,WAAW,YAAY,MAAM,6BAA6B;CAChE,IAAI,UACF,OAAO;EAAE,QAAQ,SAAS;EAAI,SAAS,GAAG,SAAS,GAAG,GAAG,SAAS;EAAM;CAI1E,MAAM,WAAW,YAAY,MAAM,6BAA6B;CAChE,IAAI,UACF,OAAO;EAAE,QAAQ,SAAS;EAAI,SAAS,GAAG,SAAS,GAAG,GAAG,SAAS;EAAM;CAI1E,MAAM,WAAW,YAAY,MAAM,uBAAuB;CAC1D,IAAI,UACF,OAAO;EAAE,QAAQ,SAAS;EAAI,SAAS,SAAS;EAAI;CAItD,MAAM,WAAW,YAAY,MAAM,uBAAuB;CAC1D,IAAI,UACF,OAAO;EAAE,QAAQ,SAAS;EAAI,SAAS,SAAS;EAAI"}

package/dist/{server-BgkyqOM7.js → server-B15THY2f.js}

@@ -1,6 +1,7 @@
 import { C as resolveEffectiveProviderType, E as setModelMappings, O as PATHS, T as resolveProviderAuthType, _ as isResponsesApiWebSearchEnabled, a as getExtraPromptForModel, b as listEnabledProviders, c as getModelResponsesApiCompactThreshold$1, d as getReasoningEffortForModel, f as getSmallModel, g as isMessagesApiEnabled, h as isGpt56OrAbove, i as getConfig, l as getProviderConfig, m as isContextManagementEnabledForResponses, n as getAnthropicApiKey, o as getMessageApiWebSearchModel, p as isContextManagementEnabledForMessages, r as getClaudeTokenMultiplier, s as getModelMappings, u as getRawProviderConfig, v as isResponsesApiWebSocketEnabled, w as resolveMappedModel } from "./config-DkgRPYZ6.js";
 import { B as HTTPError, C as prepareMessageProxyHeaders, E as compactMessageSections, F as requestContext, I as resolveTraceId$1, L as createPooledWebSocketStream, M as buildCodexRequestHeaders, N as forwardCodexResponses, O as compactSystemPromptStarts, P as generateTraceId, R as createWebSocketUrl, S as prepareInteractionHeaders, T as compactAutoContinuePromptStarts, V as forwardError, b as copilotWebSocketHeaders, d as generateRequestIdFromPayload, f as getRootSessionId, g as getCopilotUsage, h as parseUserIdMetadata, j as CODEX_API_BASE_URL, m as isNullish, p as getUUID, r as setupCodexToken, v as copilotBaseUrl, x as prepareForCompact, y as copilotHeaders, z as state } from "./token-DHWoZfxf.js";
 import { a as isDeferredToolName, c as parseMcpToolSearchSentinel, d as shouldEnableResponsesToolSearch, i as isBridgeToolSearchName, l as resolveBridgeToolSearchName, o as listDeferredToolNames, r as formatToolSearchBridgeArguments, s as normalizeToolSearchBridgeArguments, t as BRIDGE_TOOL_SEARCH_NAME, u as selectDeferredToolsByNames } from "./tool-search-OX6iPJ9D.js";
+import { i as toClientModelId, r as normalizeSdkModelId, t as findEndpointModel } from "./models-haiuFowF.js";
 import consola from "consola";
 import { createHash } from "node:crypto";
 import fs, { readFileSync } from "node:fs";
@@ -1335,6 +1336,16 @@ const BUILTIN_PROVIDER_PRICING = {
 			input: 1.4,
 			output: 4.4
 		},
+		"grok-4.5": { tiers: [{
+			cachedInput: .5,
+			input: 2,
+			maxInputTokens: 2e5,
+			output: 6
+		}, {
+			cachedInput: 1,
+			input: 4,
+			output: 12
+		}] },
 		"deepseek-v4-flash": {
 			cachedInput: .0028,
 			input: .14,
@@ -1350,6 +1361,11 @@ const BUILTIN_PROVIDER_PRICING = {
 			input: .95,
 			output: 4
 		},
+		"kimi-k3": {
+			cachedInput: .3,
+			input: 3,
+			output: 15
+		},
 		"mimo-v2.5": {
 			cachedInput: .0028,
 			input: .14,
@@ -2600,72 +2616,9 @@ function getAnthropicToolUseBlocks(toolCalls) {
 		input: JSON.parse(toolCall.function.arguments)
 	}));
 }
-//#endregion
-//#region src/lib/models.ts
-/**
-* Converts a Copilot upstream model ID to a client-friendly ID that Claude Code
-* and Claude Desktop recognize (dots in version replaced with hyphens).
-* e.g. "claude-sonnet-4.6" -> "claude-sonnet-4-6"
-* Non-Claude models are returned unchanged.
-*/
-const toClientModelId = (modelId) => {
-	const normalized = normalizeSdkModelId(modelId);
-	if (!normalized) return modelId;
-	const versionHyphenated = normalized.version.replaceAll(".", "-");
-	return `claude-${normalized.family}-${versionHyphenated}`;
-};
-const findEndpointModel = (sdkModelId) => {
-	const models = state.models?.data ?? [];
-	const exactMatch = models.find((m) => m.id === sdkModelId);
-	if (exactMatch) return exactMatch;
-	const normalized = normalizeSdkModelId(sdkModelId);
-	if (!normalized) return;
-	const modelName = `claude-${normalized.family}-${normalized.version}`;
-	const model = models.find((m) => m.id === modelName);
-	if (model) return model;
-};
-/**
-* Normalizes an SDK model ID to extract the model family and version.
-* this method from github copilot extension
-* Examples:
-* - "claude-opus-4-5-20251101" -> { family: "opus", version: "4.5" }
-* - "claude-3-5-sonnet-20241022" -> { family: "sonnet", version: "3.5" }
-* - "claude-sonnet-4-20250514" -> { family: "sonnet", version: "4" }
-* - "claude-haiku-3-5-20250514" -> { family: "haiku", version: "3.5" }
-* - "claude-haiku-4.5" -> { family: "haiku", version: "4.5" }
-*/
-const normalizeSdkModelId = (sdkModelId) => {
-	const withoutDate = sdkModelId.toLowerCase().replace(/-\d{8}$/, "");
-	const pattern1 = withoutDate.match(/^claude-(\w+)-(\d+)\.(\d+)$/);
-	if (pattern1) return {
-		family: pattern1[1],
-		version: `${pattern1[2]}.${pattern1[3]}`
-	};
-	const pattern2 = withoutDate.match(/^claude-(\w+)-(\d+)-(\d+)$/);
-	if (pattern2) return {
-		family: pattern2[1],
-		version: `${pattern2[2]}.${pattern2[3]}`
-	};
-	const pattern3 = withoutDate.match(/^claude-(\d+)-(\d+)-(\w+)$/);
-	if (pattern3) return {
-		family: pattern3[3],
-		version: `${pattern3[1]}.${pattern3[2]}`
-	};
-	const pattern4 = withoutDate.match(/^claude-(\w+)-(\d+)$/);
-	if (pattern4) return {
-		family: pattern4[1],
-		version: pattern4[2]
-	};
-	const pattern5 = withoutDate.match(/^claude-(\d+)-(\w+)$/);
-	if (pattern5) return {
-		family: pattern5[2],
-		version: pattern5[1]
-	};
-};
 const SYSTEM_REMINDER_START = "<system-reminder>";
 const SYSTEM_REMINDER_END = "</system-reminder>";
 const SUBAGENT_START_HOOK_ADDITIONAL_PREFIX = "SubagentStart hook additional";
-const IDE_EXECUTE_CODE_TOOL = "mcp__ide__executeCode";
 const IDE_GET_DIAGNOSTICS_TOOL = "mcp__ide__getDiagnostics";
 const IDE_GET_DIAGNOSTICS_DESCRIPTION = "Get language diagnostics from VS Code. Returns errors, warnings, information, and hints for files in the workspace.";
 const PDF_FILE_READ_PREFIX = "PDF file read:";
@@ -2988,7 +2941,6 @@ const mergeToolResultForClaude = (anthropicPayload, options) => {
 const sanitizeIdeTools = (payload) => {
 	if (!payload.tools || payload.tools.length === 0) return;
 	payload.tools = payload.tools.flatMap((tool) => {
-		if (tool.name === IDE_EXECUTE_CODE_TOOL && !tool.defer_loading) return [];
 		if (tool.name === IDE_GET_DIAGNOSTICS_TOOL) return [{
 			...tool,
 			description: IDE_GET_DIAGNOSTICS_DESCRIPTION
@@ -7154,4 +7106,4 @@ server.route("/:provider/images", providerImageRoutes);
 //#endregion
 export { server };
 
-//# sourceMappingURL=server-BgkyqOM7.js.map
+//# sourceMappingURL=server-B15THY2f.js.map