claudish 5.4.1 → 5.5.0

package/dist/index.js

@@ -27321,7 +27321,7 @@ Tokens: ${result.usage.input} input, ${result.usage.output} output`;
       };
     }
   });
-  server.tool("list_models", "List recommended OpenRouter models for coding tasks", {}, async () => {
+  server.tool("list_models", "List recommended models for coding tasks", {}, async () => {
     const models = loadRecommendedModels();
     if (models.length === 0) {
       return {
@@ -27347,12 +27347,12 @@ Tokens: ${result.usage.input} input, ${result.usage.output} output`;
     output += `
 ## Quick Picks
 `;
-    output += "- **Fast & cheap**: `x-ai/grok-code-fast-1` ($0.85/1M)\n";
-    output += "- **Most advanced**: `openai/gpt-5.3` ($5.00/1M)\n";
-    output += "- **Large context**: `google/gemini-3-pro-preview` (1M tokens)\n";
-    output += "- **Reasoning**: `moonshotai/kimi-k2-thinking` ($2.50/1M)\n";
-    output += "- **Budget**: `minimax/minimax-m2.1` ($0.60/1M)\n";
-    output += "- **Code specialist**: `deepseek/deepseek-v3.2` ($1.00/1M)\n";
+    output += "- **Budget**: `minimax-m2.5` ($0.75/1M)\n";
+    output += "- **Large context**: `gemini-3.1-pro-preview` (1M tokens)\n";
+    output += "- **Most advanced**: `gpt-5.2` ($7.88/1M)\n";
+    output += "- **Vision + coding**: `kimi-k2.5` ($1.32/1M)\n";
+    output += "- **Agentic**: `glm-5` ($1.68/1M)\n";
+    output += "- **Multimodal**: `qwen3.5-plus-02-15` ($1.40/1M)\n";
     return { content: [{ type: "text", text: output }] };
   });
   server.tool("search_models", "Search all OpenRouter models by name, provider, or capability", {
@@ -31119,8 +31119,7 @@ function loadRecommendedModels2() {
       const data = JSON.parse(content);
       return (data.models || []).map((model) => ({
         ...model,
-        id: model.id.startsWith("openrouter@") ? model.id : `openrouter@${model.id}`,
-        source: "OpenRouter"
+        source: "Recommended"
       }));
     } catch {
       return [];
@@ -33332,6 +33331,7 @@ var init_model_parser = __esm(() => {
     { pattern: /^meta-llama\//i, provider: "ollamacloud" },
     { pattern: /^llama-/i, provider: "ollamacloud" },
     { pattern: /^llama3/i, provider: "ollamacloud" },
+    { pattern: /^qwen/i, provider: "qwen" },
     { pattern: /^poe:/i, provider: "poe" },
     { pattern: /^anthropic\//i, provider: "native-anthropic" },
     { pattern: /^claude-/i, provider: "native-anthropic" }
@@ -33995,12 +33995,14 @@ var init_auto_route = __esm(() => {
   OPENROUTER_VENDOR_MAP = {
     google: "google",
     openai: "openai",
-    kimi: "moonshot",
-    "kimi-coding": "moonshot",
-    glm: "zhipuai",
-    "glm-coding": "zhipuai",
+    kimi: "moonshotai",
+    "kimi-coding": "moonshotai",
+    glm: "z-ai",
+    "glm-coding": "z-ai",
+    zai: "z-ai",
     minimax: "minimax",
-    ollamacloud: "meta-llama"
+    ollamacloud: "meta-llama",
+    qwen: "qwen"
   };
   PROVIDER_HINT_MAP = {
     "kimi-coding": {
@@ -35686,24 +35688,20 @@ function printAvailableModels() {
     const modelInfo = loadModelInfo();
     for (const model of basicModels) {
       const info = modelInfo[model];
-      const displayModel = model === "custom" ? model : `openrouter@${model}`;
-      console.log(`  ${displayModel}`);
+      console.log(`  ${model}`);
       console.log(`    ${info.name} - ${info.description}`);
       console.log("");
     }
     return;
   }
   console.log(`
-Available OpenRouter Models (last updated: ${lastUpdated}):
+Recommended Models (last updated: ${lastUpdated}):
 `);
-  console.log("  Model                          Provider    Pricing     Context  Capabilities");
-  console.log("  " + "─".repeat(86));
+  console.log("  Model                        Pricing     Context  Capabilities");
+  console.log("  " + "─".repeat(66));
   for (const model of models) {
-    const fullModelId = model.id.startsWith("openrouter@") ? model.id : `openrouter@${model.id}`;
-    const modelId = fullModelId.length > 30 ? fullModelId.substring(0, 27) + "..." : fullModelId;
-    const modelIdPadded = modelId.padEnd(30);
-    const provider = model.provider.length > 10 ? model.provider.substring(0, 7) + "..." : model.provider;
-    const providerPadded = provider.padEnd(10);
+    const modelId = model.id.length > 28 ? model.id.substring(0, 25) + "..." : model.id;
+    const modelIdPadded = modelId.padEnd(28);
     let pricing = model.pricing?.average || "N/A";
     if (pricing.includes("-1000000")) {
       pricing = "varies";
@@ -35717,7 +35715,7 @@ Available OpenRouter Models (last updated: ${lastUpdated}):
     const reasoning = model.supportsReasoning ? "\uD83E\uDDE0" : "  ";
     const vision = model.supportsVision ? "\uD83D\uDC41️ " : "  ";
     const capabilities = `${tools} ${reasoning} ${vision}`;
-    console.log(`  ${modelIdPadded} ${providerPadded} ${pricingPadded} ${contextPadded} ${capabilities}`);
+    console.log(`  ${modelIdPadded} ${pricingPadded} ${contextPadded} ${capabilities}`);
   }
   console.log("");
   console.log("  Capabilities: \uD83D\uDD27 Tools  \uD83E\uDDE0 Reasoning  \uD83D\uDC41️  Vision");
@@ -35734,14 +35732,7 @@ function printAvailableModelsJSON() {
   try {
     const jsonContent = readFileSync9(jsonPath, "utf-8");
     const data = JSON.parse(jsonContent);
-    const outputData = {
-      ...data,
-      models: data.models.map((model) => ({
-        ...model,
-        id: model.id.startsWith("openrouter@") ? model.id : `openrouter@${model.id}`
-      }))
-    };
-    console.log(JSON.stringify(outputData, null, 2));
+    console.log(JSON.stringify(data, null, 2));
   } catch (error46) {
     const models = getAvailableModels();
     const modelInfo = loadModelInfo();
@@ -35752,7 +35743,7 @@ function printAvailableModelsJSON() {
       models: models.filter((m) => m !== "custom").map((modelId) => {
         const info = modelInfo[modelId];
         return {
-          id: `openrouter@${modelId}`,
+          id: modelId,
           name: info.name,
           description: info.description,
           provider: info.provider,
@@ -35824,7 +35815,7 @@ async function fetchGLMCodingModels2() {
     return [];
   }
 }
-var __filename5, __dirname5, VERSION = "5.4.1", CACHE_MAX_AGE_DAYS3 = 2, MODELS_JSON_PATH, CLAUDISH_CACHE_DIR3, ALL_MODELS_JSON_PATH2;
+var __filename5, __dirname5, VERSION = "5.5.0", CACHE_MAX_AGE_DAYS3 = 2, MODELS_JSON_PATH, CLAUDISH_CACHE_DIR3, ALL_MODELS_JSON_PATH2;
 var init_cli = __esm(() => {
   init_config();
   init_model_loader();
@@ -40897,6 +40888,56 @@ var init_deepseek_adapter = __esm(() => {
   };
 });
 
+// src/adapters/glm-adapter.ts
+var GLM_CONTEXT_WINDOWS, GLM_VISION_MODELS, GLMAdapter;
+var init_glm_adapter = __esm(() => {
+  init_base_adapter();
+  init_logger();
+  GLM_CONTEXT_WINDOWS = {
+    "glm-5": 128000,
+    "glm-4-plus": 128000,
+    "glm-4-long": 1e6,
+    "glm-4-flash": 128000,
+    "glm-4": 128000,
+    "glm-3-turbo": 128000
+  };
+  GLM_VISION_MODELS = ["glm-4v", "glm-4v-plus", "glm-5"];
+  GLMAdapter = class GLMAdapter extends BaseModelAdapter {
+    processTextContent(textContent, accumulatedText) {
+      return {
+        cleanedText: textContent,
+        extractedToolCalls: [],
+        wasTransformed: false
+      };
+    }
+    prepareRequest(request, originalRequest) {
+      if (originalRequest.thinking) {
+        log(`[GLMAdapter] Stripping thinking object (not supported by GLM API)`);
+        delete request.thinking;
+      }
+      return request;
+    }
+    shouldHandle(modelId) {
+      return modelId.includes("glm-") || modelId.includes("zhipu/");
+    }
+    getName() {
+      return "GLMAdapter";
+    }
+    getContextWindow() {
+      const lower = this.modelId.toLowerCase();
+      for (const [pattern, size] of Object.entries(GLM_CONTEXT_WINDOWS)) {
+        if (lower.includes(pattern))
+          return size;
+      }
+      return 128000;
+    }
+    supportsVision() {
+      const lower = this.modelId.toLowerCase();
+      return GLM_VISION_MODELS.some((m) => lower.includes(m));
+    }
+  };
+});
+
 // src/adapters/adapter-manager.ts
 class AdapterManager {
   adapters;
@@ -40908,7 +40949,8 @@ class AdapterManager {
       new OpenAIAdapter(modelId),
       new QwenAdapter(modelId),
       new MiniMaxAdapter(modelId),
-      new DeepSeekAdapter(modelId)
+      new DeepSeekAdapter(modelId),
+      new GLMAdapter(modelId)
     ];
     this.defaultAdapter = new DefaultAdapter(modelId);
   }
@@ -40932,6 +40974,7 @@ var init_adapter_manager = __esm(() => {
   init_qwen_adapter();
   init_minimax_adapter();
   init_deepseek_adapter();
+  init_glm_adapter();
 });
 
 // src/adapters/openrouter-adapter.ts

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "claudish",
-  "version": "5.4.1",
+  "version": "5.5.0",
   "description": "Run Claude Code with any model - OpenRouter, Ollama, LM Studio & local models",
   "type": "module",
   "main": "./dist/index.js",

package/recommended-models.json

@@ -4,19 +4,20 @@
   "source": "https://openrouter.ai/models?categories=programming&fmt=cards&order=top-weekly",
   "models": [
     {
-      "id": "x-ai/grok-code-fast-1",
-      "name": "xAI: Grok Code Fast 1",
-      "description": "Grok Code Fast 1 is a speedy and economical reasoning model that excels at agentic coding. With reasoning traces visible in the response, developers can steer Grok Code for high-quality work flows.",
-      "provider": "X-ai",
-      "category": "reasoning",
+      "id": "minimax-m2.5",
+      "openrouterId": "minimax/minimax-m2.5",
+      "name": "MiniMax: MiniMax M2.5",
+      "description": "MiniMax-M2.5 is a SOTA large language model designed for real-world productivity. Trained in a diverse range of complex real-world digital working environments, M2.5 builds upon the coding expertise of M2.1 to extend into general office work, reaching fluency in generating and operating Word, Excel, and Powerpoint files, context switching between diverse software environments, and working across different agent and human teams. Scoring 80.2% on SWE-Bench Verified, 51.3% on Multi-SWE-Bench, and 76.3% on BrowseComp, M2.5 is also more token efficient than previous generations, having been trained to optimize its actions and output through planning.",
+      "provider": "Minimax",
+      "category": "programming",
       "priority": 1,
       "pricing": {
-        "input": "$0.20/1M",
-        "output": "$1.50/1M",
-        "average": "$0.85/1M"
+        "input": "$0.29/1M",
+        "output": "$1.20/1M",
+        "average": "$0.75/1M"
       },
-      "context": "256K",
-      "maxOutputTokens": 10000,
+      "context": "196K",
+      "maxOutputTokens": 196608,
       "modality": "text->text",
       "supportsTools": true,
       "supportsReasoning": true,
@@ -25,37 +26,39 @@
       "recommended": true
     },
     {
-      "id": "minimax/minimax-m2.1",
-      "name": "MiniMax: MiniMax M2.1",
-      "description": "MiniMax-M2.1 is a lightweight, state-of-the-art large language model optimized for coding, agentic workflows, and modern application development. With only 10 billion activated parameters, it delivers a major jump in real-world capability while maintaining exceptional latency, scalability, and cost efficiency.\n\nCompared to its predecessor, M2.1 delivers cleaner, more concise outputs and faster perceived response times. It shows leading multilingual coding performance across major systems and application languages, achieving 49.4% on Multi-SWE-Bench and 72.5% on SWE-Bench Multilingual, and serves as a versatile agent “brain” for IDEs, coding tools, and general-purpose assistance.\n\nTo avoid degrading this model's performance, MiniMax highly recommends preserving reasoning between turns. Learn more about using reasoning_details to pass back reasoning in our [docs](https://openrouter.ai/docs/use-cases/reasoning-tokens#preserving-reasoning-blocks).",
-      "provider": "Minimax",
-      "category": "reasoning",
+      "id": "kimi-k2.5",
+      "openrouterId": "moonshotai/kimi-k2.5",
+      "name": "MoonshotAI: Kimi K2.5",
+      "description": "Kimi K2.5 is Moonshot AI's native multimodal model, delivering state-of-the-art visual coding capability and a self-directed agent swarm paradigm. Built on Kimi K2 with continued pretraining over approximately 15T mixed visual and text tokens, it delivers strong performance in general reasoning, visual coding, and agentic tool-calling.",
+      "provider": "Moonshotai",
+      "category": "vision",
       "priority": 2,
       "pricing": {
-        "input": "$0.27/1M",
-        "output": "$0.95/1M",
-        "average": "$0.61/1M"
+        "input": "$0.45/1M",
+        "output": "$2.20/1M",
+        "average": "$1.32/1M"
       },
-      "context": "196K",
-      "maxOutputTokens": null,
-      "modality": "text->text",
+      "context": "262K",
+      "maxOutputTokens": 65535,
+      "modality": "text+image->text",
       "supportsTools": true,
       "supportsReasoning": true,
-      "supportsVision": false,
+      "supportsVision": true,
       "isModerated": false,
       "recommended": true
     },
     {
-      "id": "z-ai/glm-4.7",
-      "name": "Z.ai: GLM 4.7",
-      "description": "GLM-4.7 is Z.ai’s latest flagship model, featuring upgrades in two key areas: enhanced programming capabilities and more stable multi-step reasoning/execution. It demonstrates significant improvements in executing complex agent tasks while delivering more natural conversational experiences and superior front-end aesthetics.",
+      "id": "glm-5",
+      "openrouterId": "z-ai/glm-5",
+      "name": "Z.ai: GLM 5",
+      "description": "GLM-5 is Z.ai’s flagship open-source foundation model engineered for complex systems design and long-horizon agent workflows. Built for expert developers, it delivers production-grade performance on large-scale programming tasks, rivaling leading closed-source models. With advanced agentic planning, deep backend reasoning, and iterative self-correction, GLM-5 moves beyond code generation to full-system construction and autonomous execution.",
       "provider": "Z-ai",
       "category": "reasoning",
       "priority": 3,
       "pricing": {
-        "input": "$0.30/1M",
-        "output": "$1.40/1M",
-        "average": "$0.85/1M"
+        "input": "$0.80/1M",
+        "output": "$2.56/1M",
+        "average": "$1.68/1M"
       },
       "context": "202K",
       "maxOutputTokens": null,
@@ -67,9 +70,10 @@
       "recommended": true
     },
     {
-      "id": "google/gemini-3-pro-preview",
-      "name": "Google: Gemini 3 Pro Preview",
-      "description": "Gemini 3 Pro is Google’s flagship frontier model for high-precision multimodal reasoning, combining strong performance across text, image, video, audio, and code with a 1M-token context window. Reasoning Details must be preserved when using multi-turn tool calling, see our docs here: https://openrouter.ai/docs/use-cases/reasoning-tokens#preserving-reasoning-blocks. It delivers state-of-the-art benchmark results in general reasoning, STEM problem solving, factual QA, and multimodal understanding, including leading scores on LMArena, GPQA Diamond, MathArena Apex, MMMU-Pro, and Video-MMMU. Interactions emphasize depth and interpretability: the model is designed to infer intent with minimal prompting and produce direct, insight-focused responses.\n\nBuilt for advanced development and agentic workflows, Gemini 3 Pro provides robust tool-calling, long-horizon planning stability, and strong zero-shot generation for complex UI, visualization, and coding tasks. It excels at agentic coding (SWE-Bench Verified, Terminal-Bench 2.0), multimodal analysis, and structured long-form tasks such as research synthesis, planning, and interactive learning experiences. Suitable applications include autonomous agents, coding assistants, multimodal analytics, scientific reasoning, and high-context information processing.",
+      "id": "gemini-3.1-pro-preview",
+      "openrouterId": "google/gemini-3.1-pro-preview",
+      "name": "Google: Gemini 3.1 Pro Preview",
+      "description": "Gemini 3.1 Pro Preview is Google’s frontier reasoning model, delivering enhanced software engineering performance, improved agentic reliability, and more efficient token usage across complex workflows. Building on the multimodal foundation of the Gemini 3 series, it combines high-precision reasoning across text, image, video, audio, and code with a 1M-token context window. Reasoning Details must be preserved when using multi-turn tool calling, see our docs here: https://openrouter.ai/docs/use-cases/reasoning-tokens#preserving-reasoning. The 3.1 update introduces measurable gains in SWE benchmarks and real-world coding environments, along with stronger autonomous task execution in structured domains such as finance and spreadsheet-based workflows.\n\nDesigned for advanced development and agentic systems, Gemini 3.1 Pro Preview improves long-horizon stability and tool orchestration while increasing token efficiency. It introduces a new medium thinking level to better balance cost, speed, and performance. The model excels in agentic coding, structured planning, multimodal analysis, and workflow automation, making it well-suited for autonomous agents, financial modeling, spreadsheet automation, and high-context enterprise tasks.",
       "provider": "Google",
       "category": "vision",
       "priority": 4,
@@ -88,62 +92,43 @@
       "recommended": true
     },
     {
-      "id": "moonshotai/kimi-k2-thinking",
-      "name": "MoonshotAI: Kimi K2 Thinking",
-      "description": "Kimi K2 Thinking is Moonshot AI’s most advanced open reasoning model to date, extending the K2 series into agentic, long-horizon reasoning. Built on the trillion-parameter Mixture-of-Experts (MoE) architecture introduced in Kimi K2, it activates 32 billion parameters per forward pass and supports 256 k-token context windows. The model is optimized for persistent step-by-step thought, dynamic tool invocation, and complex reasoning workflows that span hundreds of turns. It interleaves step-by-step reasoning with tool use, enabling autonomous research, coding, and writing that can persist for hundreds of sequential actions without drift.\n\nIt sets new open-source benchmarks on HLE, BrowseComp, SWE-Multilingual, and LiveCodeBench, while maintaining stable multi-agent behavior through 200–300 tool calls. Built on a large-scale MoE architecture with MuonClip optimization, it combines strong reasoning depth with high inference efficiency for demanding agentic and analytical tasks.",
-      "provider": "Moonshotai",
+      "id": "gpt-5.2",
+      "openrouterId": "openai/gpt-5.2",
+      "name": "OpenAI: GPT-5.2",
+      "description": "GPT-5.2 is the latest frontier-grade model in the GPT-5 series, offering stronger agentic and long context perfomance compared to GPT-5.1. It uses adaptive reasoning to allocate computation dynamically, responding quickly to simple queries while spending more depth on complex tasks.\n\nBuilt for broad task coverage, GPT-5.2 delivers consistent gains across math, coding, sciende, and tool calling workloads, with more coherent long-form answers and improved tool-use reliability.",
+      "provider": "Openai",
       "category": "reasoning",
       "priority": 5,
       "pricing": {
-        "input": "$0.47/1M",
-        "output": "$2.00/1M",
-        "average": "$1.24/1M"
+        "input": "$1.75/1M",
+        "output": "$14.00/1M",
+        "average": "$7.88/1M"
       },
-      "context": "131K",
-      "maxOutputTokens": null,
-      "modality": "text->text",
+      "context": "400K",
+      "maxOutputTokens": 128000,
+      "modality": "text+image+file->text",
       "supportsTools": true,
       "supportsReasoning": true,
-      "supportsVision": false,
-      "isModerated": false,
-      "recommended": true
-    },
-    {
-      "id": "deepseek/deepseek-v3.2",
-      "name": "DeepSeek: DeepSeek V3.2",
-      "description": "DeepSeek-V3.2 is a large language model designed to harmonize high computational efficiency with strong reasoning and agentic tool-use performance. It introduces DeepSeek Sparse Attention (DSA), a fine-grained sparse attention mechanism that reduces training and inference cost while preserving quality in long-context scenarios. A scalable reinforcement learning post-training framework further improves reasoning, with reported performance in the GPT-5 class, and the model has demonstrated gold-medal results on the 2025 IMO and IOI. V3.2 also uses a large-scale agentic task synthesis pipeline to better integrate reasoning into tool-use settings, boosting compliance and generalization in interactive environments.\n\nUsers can control the reasoning behaviour with the `reasoning` `enabled` boolean. [Learn more in our docs](https://openrouter.ai/docs/use-cases/reasoning-tokens#enable-reasoning-with-default-config)",
-      "provider": "Deepseek",
-      "category": "reasoning",
-      "priority": 6,
-      "pricing": {
-        "input": "$0.25/1M",
-        "output": "$0.40/1M",
-        "average": "$0.32/1M"
-      },
-      "context": "163K",
-      "maxOutputTokens": 65536,
-      "modality": "text->text",
-      "supportsTools": true,
-      "supportsReasoning": true,
-      "supportsVision": false,
-      "isModerated": false,
+      "supportsVision": true,
+      "isModerated": true,
       "recommended": true
     },
     {
-      "id": "qwen/qwen3-vl-235b-a22b-thinking",
-      "name": "Qwen: Qwen3 VL 235B A22B Thinking",
-      "description": "Qwen3-VL-235B-A22B Thinking is a multimodal model that unifies strong text generation with visual understanding across images and video. The Thinking model is optimized for multimodal reasoning in STEM and math. The series emphasizes robust perception (recognition of diverse real-world and synthetic categories), spatial understanding (2D/3D grounding), and long-form visual comprehension, with competitive results on public multimodal benchmarks for both perception and reasoning.\n\nBeyond analysis, Qwen3-VL supports agentic interaction and tool use: it can follow complex instructions over multi-image, multi-turn dialogues; align text to video timelines for precise temporal queries; and operate GUI elements for automation tasks. The models also enable visual coding workflows, turning sketches or mockups into code and assisting with UI debugging, while maintaining strong text-only performance comparable to the flagship Qwen3 language models. This makes Qwen3-VL suitable for production scenarios spanning document AI, multilingual OCR, software/UI assistance, spatial/embodied tasks, and research on vision-language agents.",
+      "id": "qwen3.5-plus-02-15",
+      "openrouterId": "qwen/qwen3.5-plus-02-15",
+      "name": "Qwen: Qwen3.5 Plus 2026-02-15",
+      "description": "The Qwen3.5 native vision-language series Plus models are built on a hybrid architecture that integrates linear attention mechanisms with sparse mixture-of-experts models, achieving higher inference efficiency. In a variety of task evaluations, the 3.5 series consistently demonstrates performance on par with state-of-the-art leading models. Compared to the 3 series, these models show a leap forward in both pure-text and multimodal capabilities.",
       "provider": "Qwen",
       "category": "vision",
-      "priority": 7,
+      "priority": 6,
       "pricing": {
-        "input": "FREE",
-        "output": "FREE",
-        "average": "FREE"
+        "input": "$0.40/1M",
+        "output": "$2.40/1M",
+        "average": "$1.40/1M"
       },
-      "context": "131K",
-      "maxOutputTokens": 32768,
-      "modality": "text+image->text",
+      "context": "1000K",
+      "maxOutputTokens": 65536,
+      "modality": "text+image+video->text",
       "supportsTools": true,
       "supportsReasoning": true,
       "supportsVision": true,