@blockrun/clawrouter 0.9.39 → 0.10.0

package/dist/index.d.ts

@@ -683,6 +683,8 @@ declare function resolveModelAlias(model: string): string;
 type BlockRunModel = {
     id: string;
     name: string;
+    /** Model version (e.g., "4.6", "3.1", "5.2") for tracking updates */
+    version?: string;
     inputPrice: number;
     outputPrice: number;
     contextWindow: number;

package/dist/index.js

@@ -1,34 +1,36 @@
 // src/models.ts
 var MODEL_ALIASES = {
-  // Claude - short names (backend uses bare model names without anthropic/ prefix)
-  claude: "claude-sonnet-4",
-  sonnet: "claude-sonnet-4",
-  "sonnet-4.6": "claude-sonnet-4",
-  "sonnet-4-6": "claude-sonnet-4",
-  opus: "claude-opus-4",
-  "opus-4": "claude-opus-4",
-  haiku: "claude-haiku-4.5",
+  // Claude - use newest versions (4.6)
+  claude: "anthropic/claude-sonnet-4.6",
+  sonnet: "anthropic/claude-sonnet-4.6",
+  "sonnet-4": "anthropic/claude-sonnet-4.6",
+  "sonnet-4.6": "anthropic/claude-sonnet-4.6",
+  "sonnet-4-6": "anthropic/claude-sonnet-4.6",
+  opus: "anthropic/claude-opus-4.6",
+  "opus-4": "anthropic/claude-opus-4.6",
+  "opus-4.6": "anthropic/claude-opus-4.6",
+  "opus-4-6": "anthropic/claude-opus-4.6",
+  haiku: "anthropic/claude-haiku-4.5",
   // Claude - provider/shortname patterns (common in agent frameworks)
-  "anthropic/sonnet": "claude-sonnet-4",
-  "anthropic/opus": "claude-opus-4",
-  "anthropic/haiku": "claude-haiku-4.5",
-  "anthropic/claude": "claude-sonnet-4",
-  // Backward compatibility - various formats all route to backend names
-  "anthropic/claude-sonnet-4": "claude-sonnet-4",
-  "anthropic/claude-sonnet-4-6": "claude-sonnet-4",
-  "anthropic/claude-sonnet-4.6": "claude-sonnet-4",
-  "anthropic/claude-opus-4": "claude-opus-4",
-  "anthropic/claude-opus-4-6": "claude-opus-4",
-  "anthropic/claude-opus-4.6": "claude-opus-4",
-  "anthropic/claude-haiku-4": "claude-haiku-4.5",
-  "anthropic/claude-haiku-4-5": "claude-haiku-4.5",
-  "anthropic/claude-haiku-4.5": "claude-haiku-4.5",
+  "anthropic/sonnet": "anthropic/claude-sonnet-4.6",
+  "anthropic/opus": "anthropic/claude-opus-4.6",
+  "anthropic/haiku": "anthropic/claude-haiku-4.5",
+  "anthropic/claude": "anthropic/claude-sonnet-4.6",
+  // Backward compatibility - map all variants to 4.6
+  "anthropic/claude-sonnet-4": "anthropic/claude-sonnet-4.6",
+  "anthropic/claude-sonnet-4-6": "anthropic/claude-sonnet-4.6",
+  "anthropic/claude-opus-4": "anthropic/claude-opus-4.6",
+  "anthropic/claude-opus-4-6": "anthropic/claude-opus-4.6",
+  "anthropic/claude-opus-4.5": "anthropic/claude-opus-4.6",
+  "anthropic/claude-haiku-4": "anthropic/claude-haiku-4.5",
+  "anthropic/claude-haiku-4-5": "anthropic/claude-haiku-4.5",
   // OpenAI
   gpt: "openai/gpt-4o",
   gpt4: "openai/gpt-4o",
   gpt5: "openai/gpt-5.2",
   codex: "openai/gpt-5.2-codex",
   mini: "openai/gpt-4o-mini",
+  o1: "openai/o1",
   o3: "openai/o3",
   // DeepSeek
   deepseek: "deepseek/deepseek-chat",
@@ -104,6 +106,7 @@ var BLOCKRUN_MODELS = [
   {
     id: "openai/gpt-5.2",
     name: "GPT-5.2",
+    version: "5.2",
     inputPrice: 1.75,
     outputPrice: 14,
     contextWindow: 4e5,
@@ -115,6 +118,7 @@ var BLOCKRUN_MODELS = [
   {
     id: "openai/gpt-5-mini",
     name: "GPT-5 Mini",
+    version: "5.0",
     inputPrice: 0.25,
     outputPrice: 2,
     contextWindow: 2e5,
@@ -123,6 +127,7 @@ var BLOCKRUN_MODELS = [
   {
     id: "openai/gpt-5-nano",
     name: "GPT-5 Nano",
+    version: "5.0",
     inputPrice: 0.05,
     outputPrice: 0.4,
     contextWindow: 128e3,
@@ -131,6 +136,7 @@ var BLOCKRUN_MODELS = [
   {
     id: "openai/gpt-5.2-pro",
     name: "GPT-5.2 Pro",
+    version: "5.2",
     inputPrice: 21,
     outputPrice: 168,
     contextWindow: 4e5,
@@ -141,8 +147,9 @@ var BLOCKRUN_MODELS = [
   {
     id: "openai/gpt-5.2-codex",
     name: "GPT-5.2 Codex",
-    inputPrice: 2.5,
-    outputPrice: 12,
+    version: "5.2",
+    inputPrice: 1.75,
+    outputPrice: 14,
     contextWindow: 128e3,
     maxOutput: 32e3,
     agentic: true
@@ -151,6 +158,7 @@ var BLOCKRUN_MODELS = [
   {
     id: "openai/gpt-4.1",
     name: "GPT-4.1",
+    version: "4.1",
     inputPrice: 2,
     outputPrice: 8,
     contextWindow: 128e3,
@@ -160,15 +168,25 @@ var BLOCKRUN_MODELS = [
   {
     id: "openai/gpt-4.1-mini",
     name: "GPT-4.1 Mini",
+    version: "4.1",
     inputPrice: 0.4,
     outputPrice: 1.6,
     contextWindow: 128e3,
     maxOutput: 16384
   },
-  // gpt-4.1-nano removed - replaced by gpt-5-nano
+  {
+    id: "openai/gpt-4.1-nano",
+    name: "GPT-4.1 Nano",
+    version: "4.1",
+    inputPrice: 0.1,
+    outputPrice: 0.4,
+    contextWindow: 128e3,
+    maxOutput: 16384
+  },
   {
     id: "openai/gpt-4o",
     name: "GPT-4o",
+    version: "4o",
     inputPrice: 2.5,
     outputPrice: 10,
     contextWindow: 128e3,
@@ -179,15 +197,37 @@ var BLOCKRUN_MODELS = [
   {
     id: "openai/gpt-4o-mini",
     name: "GPT-4o Mini",
+    version: "4o-mini",
     inputPrice: 0.15,
     outputPrice: 0.6,
     contextWindow: 128e3,
     maxOutput: 16384
   },
-  // OpenAI O-series (Reasoning) - o1/o1-mini removed, replaced by o3/o4
+  // OpenAI O-series (Reasoning)
+  {
+    id: "openai/o1",
+    name: "o1",
+    version: "1",
+    inputPrice: 15,
+    outputPrice: 60,
+    contextWindow: 2e5,
+    maxOutput: 1e5,
+    reasoning: true
+  },
+  {
+    id: "openai/o1-mini",
+    name: "o1-mini",
+    version: "1-mini",
+    inputPrice: 1.1,
+    outputPrice: 4.4,
+    contextWindow: 128e3,
+    maxOutput: 65536,
+    reasoning: true
+  },
   {
     id: "openai/o3",
     name: "o3",
+    version: "3",
     inputPrice: 2,
     outputPrice: 8,
     contextWindow: 2e5,
@@ -197,6 +237,7 @@ var BLOCKRUN_MODELS = [
   {
     id: "openai/o3-mini",
     name: "o3-mini",
+    version: "3-mini",
     inputPrice: 1.1,
     outputPrice: 4.4,
     contextWindow: 128e3,
@@ -206,6 +247,7 @@ var BLOCKRUN_MODELS = [
   {
     id: "openai/o4-mini",
     name: "o4-mini",
+    version: "4-mini",
     inputPrice: 1.1,
     outputPrice: 4.4,
     contextWindow: 128e3,
@@ -213,10 +255,11 @@ var BLOCKRUN_MODELS = [
     reasoning: true
   },
   // Anthropic - all Claude models excel at agentic workflows
-  // Backend uses bare model names (claude-sonnet-4, not anthropic/claude-sonnet-4-6)
+  // Use newest versions (4.6) with full provider prefix
   {
-    id: "claude-haiku-4.5",
+    id: "anthropic/claude-haiku-4.5",
     name: "Claude Haiku 4.5",
+    version: "4.5",
     inputPrice: 1,
     outputPrice: 5,
     contextWindow: 2e5,
@@ -224,8 +267,9 @@ var BLOCKRUN_MODELS = [
     agentic: true
   },
   {
-    id: "claude-sonnet-4",
+    id: "anthropic/claude-sonnet-4.6",
     name: "Claude Sonnet 4.6",
+    version: "4.6",
     inputPrice: 3,
     outputPrice: 15,
     contextWindow: 2e5,
@@ -234,19 +278,32 @@ var BLOCKRUN_MODELS = [
     agentic: true
   },
   {
-    id: "claude-opus-4",
+    id: "anthropic/claude-opus-4.6",
     name: "Claude Opus 4.6",
-    inputPrice: 15,
-    outputPrice: 75,
+    version: "4.6",
+    inputPrice: 5,
+    outputPrice: 25,
     contextWindow: 2e5,
     maxOutput: 32e3,
     reasoning: true,
     agentic: true
   },
   // Google
+  {
+    id: "google/gemini-3.1-pro-preview",
+    name: "Gemini 3.1 Pro Preview",
+    version: "3.1",
+    inputPrice: 2,
+    outputPrice: 12,
+    contextWindow: 105e4,
+    maxOutput: 65536,
+    reasoning: true,
+    vision: true
+  },
   {
     id: "google/gemini-3-pro-preview",
     name: "Gemini 3 Pro Preview",
+    version: "3.0",
     inputPrice: 2,
     outputPrice: 12,
     contextWindow: 105e4,
@@ -254,9 +311,20 @@ var BLOCKRUN_MODELS = [
     reasoning: true,
     vision: true
   },
+  {
+    id: "google/gemini-3-flash-preview",
+    name: "Gemini 3 Flash Preview",
+    version: "3.0",
+    inputPrice: 0.5,
+    outputPrice: 3,
+    contextWindow: 1e6,
+    maxOutput: 65536,
+    vision: true
+  },
   {
     id: "google/gemini-2.5-pro",
     name: "Gemini 2.5 Pro",
+    version: "2.5",
     inputPrice: 1.25,
     outputPrice: 10,
     contextWindow: 105e4,
@@ -267,8 +335,18 @@ var BLOCKRUN_MODELS = [
   {
     id: "google/gemini-2.5-flash",
     name: "Gemini 2.5 Flash",
-    inputPrice: 0.15,
-    outputPrice: 0.6,
+    version: "2.5",
+    inputPrice: 0.3,
+    outputPrice: 2.5,
+    contextWindow: 1e6,
+    maxOutput: 65536
+  },
+  {
+    id: "google/gemini-2.5-flash-lite",
+    name: "Gemini 2.5 Flash Lite",
+    version: "2.5",
+    inputPrice: 0.1,
+    outputPrice: 0.4,
     contextWindow: 1e6,
     maxOutput: 65536
   },
@@ -276,6 +354,7 @@ var BLOCKRUN_MODELS = [
   {
     id: "deepseek/deepseek-chat",
     name: "DeepSeek V3.2 Chat",
+    version: "3.2",
     inputPrice: 0.28,
     outputPrice: 0.42,
     contextWindow: 128e3,
@@ -284,6 +363,7 @@ var BLOCKRUN_MODELS = [
   {
     id: "deepseek/deepseek-reasoner",
     name: "DeepSeek V3.2 Reasoner",
+    version: "3.2",
     inputPrice: 0.28,
     outputPrice: 0.42,
     contextWindow: 128e3,
@@ -294,8 +374,9 @@ var BLOCKRUN_MODELS = [
   {
     id: "moonshot/kimi-k2.5",
     name: "Kimi K2.5",
-    inputPrice: 0.5,
-    outputPrice: 2.4,
+    version: "k2.5",
+    inputPrice: 0.6,
+    outputPrice: 3,
     contextWindow: 262144,
     maxOutput: 8192,
     reasoning: true,
@@ -306,6 +387,7 @@ var BLOCKRUN_MODELS = [
   {
     id: "xai/grok-3",
     name: "Grok 3",
+    version: "3",
     inputPrice: 3,
     outputPrice: 15,
     contextWindow: 131072,
@@ -316,6 +398,7 @@ var BLOCKRUN_MODELS = [
   {
     id: "xai/grok-3-mini",
     name: "Grok 3 Mini",
+    version: "3-mini",
     inputPrice: 0.3,
     outputPrice: 0.5,
     contextWindow: 131072,
@@ -325,6 +408,7 @@ var BLOCKRUN_MODELS = [
   {
     id: "xai/grok-4-fast-reasoning",
     name: "Grok 4 Fast Reasoning",
+    version: "4",
     inputPrice: 0.2,
     outputPrice: 0.5,
     contextWindow: 131072,
@@ -334,6 +418,7 @@ var BLOCKRUN_MODELS = [
   {
     id: "xai/grok-4-fast-non-reasoning",
     name: "Grok 4 Fast",
+    version: "4",
     inputPrice: 0.2,
     outputPrice: 0.5,
     contextWindow: 131072,
@@ -342,6 +427,7 @@ var BLOCKRUN_MODELS = [
   {
     id: "xai/grok-4-1-fast-reasoning",
     name: "Grok 4.1 Fast Reasoning",
+    version: "4.1",
     inputPrice: 0.2,
     outputPrice: 0.5,
     contextWindow: 131072,
@@ -351,6 +437,7 @@ var BLOCKRUN_MODELS = [
   {
     id: "xai/grok-4-1-fast-non-reasoning",
     name: "Grok 4.1 Fast",
+    version: "4.1",
     inputPrice: 0.2,
     outputPrice: 0.5,
     contextWindow: 131072,
@@ -359,6 +446,7 @@ var BLOCKRUN_MODELS = [
   {
     id: "xai/grok-code-fast-1",
     name: "Grok Code Fast",
+    version: "1",
     inputPrice: 0.2,
     outputPrice: 1.5,
     contextWindow: 131072,
@@ -369,17 +457,28 @@ var BLOCKRUN_MODELS = [
   {
     id: "xai/grok-4-0709",
     name: "Grok 4 (0709)",
+    version: "4-0709",
     inputPrice: 0.2,
     outputPrice: 1.5,
     contextWindow: 131072,
     maxOutput: 16384,
     reasoning: true
   },
-  // grok-2-vision removed - old, 0 transactions
+  {
+    id: "xai/grok-2-vision",
+    name: "Grok 2 Vision",
+    version: "2",
+    inputPrice: 2,
+    outputPrice: 10,
+    contextWindow: 131072,
+    maxOutput: 16384,
+    vision: true
+  },
   // MiniMax
   {
     id: "minimax/minimax-m2.5",
     name: "MiniMax M2.5",
+    version: "m2.5",
     inputPrice: 0.3,
     outputPrice: 1.2,
     contextWindow: 204800,
@@ -391,6 +490,7 @@ var BLOCKRUN_MODELS = [
   {
     id: "nvidia/gpt-oss-120b",
     name: "NVIDIA GPT-OSS 120B",
+    version: "120b",
     inputPrice: 0,
     outputPrice: 0,
     contextWindow: 128e3,
@@ -399,6 +499,7 @@ var BLOCKRUN_MODELS = [
   {
     id: "nvidia/kimi-k2.5",
     name: "NVIDIA Kimi K2.5",
+    version: "k2.5",
     inputPrice: 0.55,
     outputPrice: 2.5,
     contextWindow: 262144,
@@ -1681,10 +1782,10 @@ var DEFAULT_ROUTING_CONFIG = {
   tiers: {
     SIMPLE: {
       primary: "moonshot/kimi-k2.5",
-      // $0.50/$2.40 - best quality/price for simple tasks
+      // $0.60/$3.00 - best quality/price for simple tasks
       fallback: [
-        "google/gemini-2.5-flash",
-        // 1M context, cost-effective
+        "google/gemini-2.5-flash-lite",
+        // 1M context, ultra cheap ($0.10/$0.40)
         "nvidia/gpt-oss-120b",
         // FREE fallback
         "deepseek/deepseek-chat"
@@ -1694,26 +1795,28 @@ var DEFAULT_ROUTING_CONFIG = {
       primary: "xai/grok-code-fast-1",
       // Code specialist, $0.20/$1.50
       fallback: [
-        "google/gemini-2.5-flash",
-        // 1M context, cost-effective
+        "google/gemini-2.5-flash-lite",
+        // 1M context, ultra cheap ($0.10/$0.40)
         "deepseek/deepseek-chat",
         "xai/grok-4-1-fast-non-reasoning"
         // Upgraded Grok 4.1
       ]
     },
     COMPLEX: {
-      primary: "google/gemini-3-pro-preview",
-      // Latest Gemini - upgraded from 2.5
+      primary: "google/gemini-3.1-pro-preview",
+      // Newest Gemini 3.1 - upgraded from 3.0
       fallback: [
-        "google/gemini-2.5-flash",
-        // CRITICAL: 1M context, cheap failsafe before expensive models
+        "google/gemini-2.5-flash-lite",
+        // CRITICAL: 1M context, ultra-cheap failsafe ($0.10/$0.40)
+        "google/gemini-3-pro-preview",
+        // 3.0 fallback
         "google/gemini-2.5-pro",
         "deepseek/deepseek-chat",
         "xai/grok-4-0709",
         "openai/gpt-5.2",
         // Newer and cheaper input than gpt-4o
         "openai/gpt-4o",
-        "claude-sonnet-4"
+        "anthropic/claude-sonnet-4.6"
       ]
     },
     REASONING: {
@@ -1733,17 +1836,17 @@ var DEFAULT_ROUTING_CONFIG = {
     SIMPLE: {
       primary: "nvidia/gpt-oss-120b",
       // FREE! $0.00/$0.00
-      fallback: ["google/gemini-2.5-flash", "deepseek/deepseek-chat"]
+      fallback: ["google/gemini-2.5-flash-lite", "deepseek/deepseek-chat"]
     },
     MEDIUM: {
-      primary: "google/gemini-2.5-flash",
-      // $0.15/$0.60 - cheapest capable
+      primary: "google/gemini-2.5-flash-lite",
+      // $0.10/$0.40 - cheapest capable with 1M context
       fallback: ["deepseek/deepseek-chat", "nvidia/gpt-oss-120b"]
     },
     COMPLEX: {
-      primary: "google/gemini-2.5-flash",
-      // $0.15/$0.60 - 1M context handles complexity
-      fallback: ["deepseek/deepseek-chat", "xai/grok-4-0709"]
+      primary: "google/gemini-2.5-flash-lite",
+      // $0.10/$0.40 - 1M context handles complexity
+      fallback: ["google/gemini-2.5-flash", "deepseek/deepseek-chat", "xai/grok-4-0709"]
     },
     REASONING: {
       primary: "xai/grok-4-1-fast-reasoning",
@@ -1756,8 +1859,8 @@ var DEFAULT_ROUTING_CONFIG = {
   premiumTiers: {
     SIMPLE: {
       primary: "moonshot/kimi-k2.5",
-      // $0.50/$2.40 - good for simple coding
-      fallback: ["claude-haiku-4.5", "google/gemini-2.5-flash", "xai/grok-code-fast-1"]
+      // $0.60/$3.00 - good for simple coding
+      fallback: ["anthropic/claude-haiku-4.5", "google/gemini-2.5-flash-lite", "xai/grok-code-fast-1"]
     },
     MEDIUM: {
       primary: "openai/gpt-5.2-codex",
@@ -1766,26 +1869,28 @@ var DEFAULT_ROUTING_CONFIG = {
         "moonshot/kimi-k2.5",
         "google/gemini-2.5-pro",
         "xai/grok-4-0709",
-        "claude-sonnet-4"
+        "anthropic/claude-sonnet-4.6"
       ]
     },
     COMPLEX: {
-      primary: "claude-opus-4",
+      primary: "anthropic/claude-opus-4.6",
       // Best quality for complex tasks
       fallback: [
         "openai/gpt-5.2-codex",
-        "claude-opus-4",
-        "claude-sonnet-4",
+        "anthropic/claude-opus-4.6",
+        "anthropic/claude-sonnet-4.6",
+        "google/gemini-3.1-pro-preview",
+        // Newest Gemini
         "google/gemini-3-pro-preview",
         "moonshot/kimi-k2.5"
       ]
     },
     REASONING: {
-      primary: "claude-sonnet-4",
+      primary: "anthropic/claude-sonnet-4.6",
       // $3/$15 - best for reasoning/instructions
       fallback: [
-        "claude-opus-4",
-        "claude-opus-4",
+        "anthropic/claude-opus-4.6",
+        "anthropic/claude-opus-4.6",
         "openai/o4-mini",
         // Newer and cheaper than o3 ($1.10 vs $2.00)
         "openai/o3",
@@ -1798,27 +1903,29 @@ var DEFAULT_ROUTING_CONFIG = {
     SIMPLE: {
       primary: "moonshot/kimi-k2.5",
       // Cheaper than Haiku ($0.5/$2.4 vs $1/$5), larger context
-      fallback: ["claude-haiku-4.5", "xai/grok-4-1-fast-non-reasoning", "openai/gpt-4o-mini"]
+      fallback: ["anthropic/claude-haiku-4.5", "xai/grok-4-1-fast-non-reasoning", "openai/gpt-4o-mini"]
     },
     MEDIUM: {
       primary: "xai/grok-code-fast-1",
       // Code specialist for agentic coding
-      fallback: ["moonshot/kimi-k2.5", "claude-haiku-4.5", "claude-sonnet-4"]
+      fallback: ["moonshot/kimi-k2.5", "anthropic/claude-haiku-4.5", "claude-sonnet-4"]
     },
     COMPLEX: {
-      primary: "claude-sonnet-4",
+      primary: "anthropic/claude-sonnet-4.6",
       fallback: [
-        "claude-opus-4",
+        "anthropic/claude-opus-4.6",
         // Latest Opus - best agentic
         "openai/gpt-5.2",
+        "google/gemini-3.1-pro-preview",
+        // Newest Gemini
         "google/gemini-3-pro-preview",
         "xai/grok-4-0709"
       ]
     },
     REASONING: {
-      primary: "claude-sonnet-4",
+      primary: "anthropic/claude-sonnet-4.6",
       // Strong tool use + reasoning for agentic tasks
-      fallback: ["claude-opus-4", "xai/grok-4-1-fast-reasoning", "deepseek/deepseek-reasoner"]
+      fallback: ["anthropic/claude-opus-4.6", "xai/grok-4-1-fast-reasoning", "deepseek/deepseek-reasoner"]
     }
   },
   overrides: {