@blockrun/clawrouter 0.12.244 → 0.12.246

package/dist/router/index.js

@@ -363,6 +363,21 @@ function applyPromotions(tierConfigs, promotions, profile, now = /* @__PURE__ */
   }
   return result;
 }
+function applyUnavailableModels(tierConfigs, unavailableModels) {
+  if (!unavailableModels || unavailableModels.length === 0) return tierConfigs;
+  const dead = new Set(unavailableModels);
+  let result = tierConfigs;
+  for (const tier of Object.keys(tierConfigs)) {
+    const config = tierConfigs[tier];
+    const alive = [config.primary, ...config.fallback].filter((model) => !dead.has(model));
+    if (alive.length === 0 || alive[0] === config.primary && alive.length === config.fallback.length + 1) {
+      continue;
+    }
+    if (result === tierConfigs) result = { ...tierConfigs };
+    result[tier] = { primary: alive[0], fallback: alive.slice(1) };
+  }
+  return result;
+}
 var RulesStrategy = class {
   name = "rules";
   route(prompt, systemPrompt, maxOutputTokens, options) {
@@ -414,6 +429,7 @@ ${value.slice(-(scanLimit - prefixLength))}`;
       profile = useAgenticTiers ? "agentic" : "auto";
     }
     tierConfigs = applyPromotions(tierConfigs, config.promotions, profile, options.now);
+    tierConfigs = applyUnavailableModels(tierConfigs, options.unavailableModels);
     const agenticScoreValue = ruleResult.agenticScore;
     if (estimatedTokens > config.overrides.maxTokensForceComplex) {
       const decision2 = selectModel(
@@ -552,18 +568,6 @@ var DEFAULT_MODEL_CAPABILITIES = Object.freeze({
     supportsTools: false,
     supportsVision: false
   },
-  "free/gpt-oss-120b": {
-    contextWindow: 128e3,
-    maxOutputTokens: 16384,
-    supportsTools: false,
-    supportsVision: false
-  },
-  "free/gpt-oss-20b": {
-    contextWindow: 128e3,
-    maxOutputTokens: 16384,
-    supportsTools: false,
-    supportsVision: false
-  },
   "free/seed-oss-36b": {
     contextWindow: 131072,
     maxOutputTokens: 16384,
@@ -636,6 +640,18 @@ var DEFAULT_MODEL_CAPABILITIES = Object.freeze({
     supportsTools: true,
     supportsVision: true
   },
+  "nvidia/nemotron-nano-9b-v2": {
+    contextWindow: 131072,
+    maxOutputTokens: 16384,
+    supportsTools: false,
+    supportsVision: false
+  },
+  "nvidia/step-3.7-flash": {
+    contextWindow: 131072,
+    maxOutputTokens: 16384,
+    supportsTools: false,
+    supportsVision: false
+  },
   "openai/gpt-4.1": {
     contextWindow: 128e3,
     maxOutputTokens: 16384,
@@ -1732,9 +1748,12 @@ var PortfolioStrategy = class {
     const targetTier = (features.taskType === "reasoning_mcq" || features.taskType === "reasoning_math") && (base.tier === "SIMPLE" || base.tier === "MEDIUM") ? "REASONING" : base.tier;
     const tierConfig = tierConfigs[targetTier];
     const configuredCandidates = tierConfig ? getFallbackChain(targetTier, tierConfigs) : [];
+    const unavailable = new Set(options.unavailableModels ?? []);
     const chain = [
       .../* @__PURE__ */ new Set([...configuredCandidates, ...evidenceCandidates(features.taskType)])
-    ].filter((model2) => typeof model2 === "string" && model2.length > 0);
+    ].filter(
+      (model2) => typeof model2 === "string" && model2.length > 0 && !unavailable.has(model2)
+    );
     const eligible = chain.filter(
       (model2) => isEligible(model2, features, maxOutputTokens, options)
     );
@@ -1811,13 +1830,7 @@ var PortfolioStrategy = class {
       ...eligibleCandidates.filter(
         (model2) => !scoredModels.includes(model2) && !webResearchFallbackOrder.includes(model2)
       )
-    ] : features.taskType === "tool_agent" || features.taskType === "tool_agent_parallel" && features.agentDomain !== "other" ? [
-      ...scoredModels,
-      ...eligibleCandidates.filter((model2) => !scoredModels.includes(model2))
-    ] : [
-      ...scoredModels,
-      ...eligibleCandidates.filter((model2) => !scoredModels.includes(model2))
-    ];
+    ] : [...scoredModels, ...eligibleCandidates.filter((model2) => !scoredModels.includes(model2))];
     const model = ranked[0] ?? base.model;
     const selectedTierConfigs = {
       ...tierConfigs,
@@ -2929,8 +2942,8 @@ var DEFAULT_ROUTING_CONFIG = {
         // $0.20/$1.25, 1M context
         "xai/grok-4-fast-non-reasoning",
         // 1,143ms, $0.20/$0.50 — fast fallback
-        "free/gpt-oss-120b"
-        // 1,252ms, FREE fallback (hidden from /v1/models but direct calls work)
+        "nvidia/step-3.7-flash"
+        // FREE backstop — new NVIDIA free tier (gpt-oss-120b now 400s; probed 2026-08-21)
       ]
     },
     MEDIUM: {
@@ -3003,15 +3016,15 @@ var DEFAULT_ROUTING_CONFIG = {
   // Eco tier configs - absolute cheapest (blockrun/eco)
   ecoTiers: {
     SIMPLE: {
-      primary: "free/gpt-oss-120b",
-      // FREE! $0.00/$0.00 — heavy user default
+      primary: "nvidia/step-3.7-flash",
+      // FREE! $0.00/$0.00 — new NVIDIA free tier flagship
       fallback: [
-        "free/gpt-oss-20b",
-        // FREE — smaller, faster
-        "free/deepseek-v4-flash",
-        // FREE — 1M ctx; slow (~10 tok/s, 07-28 probe) but completes
-        // seed-oss-36b sat here as the free coder until it EOL'd 2026-08-03 (HTTP 410).
-        // gpt-oss-120b/20b already head this chain, so the rung is dropped, not retargeted.
+        "nvidia/nemotron-nano-9b-v2",
+        // FREE — compact + fast (~0.7s), high-volume light tasks
+        // This head keeps rotting with NVIDIA's free hosting: deepseek-v4-flash
+        // (410, 2026-08-12), seed-oss-36b (410, 2026-08-03), then gpt-oss-120b/20b
+        // (400 Unknown model, probed 2026-08-21). Each retirement retargets the
+        // two free rungs to the current free tier; the paid rungs below never move.
         "google/gemini-3.1-flash-lite",
         // $0.25/$1.50 — newest flash-lite
         "openai/gpt-5.4-nano",
@@ -3128,8 +3141,8 @@ var DEFAULT_ROUTING_CONFIG = {
         "openai/gpt-5.3-codex",
         "deepseek/deepseek-chat",
         // Cheap, reliable
-        "free/gpt-oss-120b"
-        // NVIDIA free ultimate backstop (was seed-oss-36b; EOL'd 2026-08-03)
+        "nvidia/step-3.7-flash"
+        // NVIDIA free ultimate backstop (was gpt-oss-120b; 400s since ~2026-08)
       ]
     },
     REASONING: {
@@ -3218,8 +3231,8 @@ var DEFAULT_ROUTING_CONFIG = {
         // Previous flagship — 6,213ms, reliable
         "deepseek/deepseek-chat",
         // 1,431ms — cheap, reliable
-        "free/gpt-oss-120b"
-        // NVIDIA free ultimate backstop (was seed-oss-36b; EOL'd 2026-08-03)
+        "nvidia/step-3.7-flash"
+        // NVIDIA free ultimate backstop (was gpt-oss-120b; 400s since ~2026-08)
       ]
     },
     REASONING: {
@@ -3276,6 +3289,7 @@ export {
   LIVE_MODEL_PROFILES,
   PortfolioStrategy,
   RulesStrategy,
+  applyUnavailableModels,
   calculateModelCost,
   classifyByRules,
   filterByExcludeList,

package/docs/configuration.md

@@ -158,17 +158,10 @@ Response (dual-chain):
 
 ```bash
 /wallet solana    # Switch to Solana USDC payments
-/wallet base      # Switch back to Base (EVM) USDC payments
+/wallet base      # Switch to Base (EVM) USDC payments
 ```
 
-Or use the `/chain` command:
-
-```bash
-/chain solana
-/chain base
-```
-
-The selected chain is persisted across gateway restarts.
+The selected chain is persisted across gateway restarts (in `~/.openclaw/blockrun/payment-chain`). New installs default to **Solana** (persisted when the wallet is first generated); existing installs without a saved selection stay on **Base**, where their funds already live. The `CLAWROUTER_PAYMENT_CHAIN` environment variable (`solana` or `base`) overrides the persisted selection.
 
 ### Switch Wallets
 

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@blockrun/clawrouter",
-  "version": "0.12.244",
-  "description": "Smart LLM router — save 88% on inference costs. 71 models (6 free), one wallet, x402 micropayments.",
+  "version": "0.12.246",
+  "description": "Smart LLM router — save 88% on inference costs. 71 models (5 free), one wallet, x402 micropayments.",
   "type": "module",
   "main": "dist/index.js",
   "types": "dist/index.d.ts",
@@ -110,7 +110,7 @@
     }
   },
   "devDependencies": {
-    "@blockrun/router-core": "https://codeload.github.com/BlockRunAI/router-core/tar.gz/6a790ebec60161825bbc8c4093fd221006b4e5fb",
+    "@blockrun/router-core": "https://codeload.github.com/BlockRunAI/router-core/tar.gz/d7bc10ce184b3daa40bcc115b523fd629d847a52",
     "@eslint/js": "^10.0.1",
     "eslint": "^10.2.0",
     "openclaw": "^2026.7.2-beta.7",

package/skills/clawrouter/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: clawrouter
-description: Hosted-gateway LLM router — save 88% on inference costs. A local proxy that forwards each request to the blockrun.ai gateway, which routes to the cheapest capable model across 71 models from OpenAI, Anthropic, Google, DeepSeek, xAI, NVIDIA, and more. 6 free NVIDIA models included. Also exposes realtime market data (global stocks, crypto, FX, commodities), Twitter/X intelligence, prediction-market data across Polymarket, Kalshi, Limitless, Opinion, Predict.Fun, dFlow + UMA oracle resolution + wallet identity & clustering, phone-number intelligence (carrier + SIM-swap fraud detection) plus AI-powered outbound voice calls (Twilio + Bland.ai), AND the Surf unified crypto data API (84 endpoints — CEX/DEX, on-chain SQL over 80+ ClickHouse tables, 100M+ labeled wallets, prediction markets, social/CT mindshare, news, VC fund intel) as built-in agent tools. Not a local-inference tool — prompts are sent to the blockrun.ai gateway.
+description: Hosted-gateway LLM router — save 88% on inference costs. A local proxy that forwards each request to the blockrun.ai gateway, which routes to the cheapest capable model across 71 models from OpenAI, Anthropic, Google, DeepSeek, xAI, NVIDIA, and more. 5 free NVIDIA models included. Also exposes realtime market data (global stocks, crypto, FX, commodities), Twitter/X intelligence, prediction-market data across Polymarket, Kalshi, Limitless, Opinion, Predict.Fun, dFlow + UMA oracle resolution + wallet identity & clustering, phone-number intelligence (carrier + SIM-swap fraud detection) plus AI-powered outbound voice calls (Twilio + Bland.ai), AND the Surf unified crypto data API (84 endpoints — CEX/DEX, on-chain SQL over 80+ ClickHouse tables, 100M+ labeled wallets, prediction markets, social/CT mindshare, news, VC fund intel) as built-in agent tools. Not a local-inference tool — prompts are sent to the blockrun.ai gateway.
 triggers:
   - "clawrouter"
   - "claw router"
@@ -53,7 +53,7 @@ metadata:
 
 # ClawRouter
 
-Hosted-gateway LLM router that saves <!-- br:savings.autoVsBaselinePct -->88<!-- /br:savings.autoVsBaselinePct -->% on inference costs by forwarding each request to the blockrun.ai gateway, which picks the cheapest model capable of handling it across <!-- br:models.chatVisible -->71<!-- /br:models.chatVisible --> models from 9 providers (<!-- br:models.free -->6<!-- /br:models.free --> free NVIDIA models). All billing flows through one USDC wallet; you do not hold provider API keys.
+Hosted-gateway LLM router that saves <!-- br:savings.autoVsBaselinePct -->88<!-- /br:savings.autoVsBaselinePct -->% on inference costs by forwarding each request to the blockrun.ai gateway, which picks the cheapest model capable of handling it across <!-- br:models.chatVisible -->71<!-- /br:models.chatVisible --> models from 9 providers (<!-- br:models.free -->5<!-- /br:models.free --> free NVIDIA models). All billing flows through one USDC wallet; you do not hold provider API keys.
 
 **This is not a local-inference tool.** ClawRouter is a thin local proxy. Your prompts are sent over HTTPS to the blockrun.ai gateway for model execution. If your workload requires inference that never leaves your machine, use a local runtime like Ollama — ClawRouter is not the right tool for that use case.
 
@@ -85,7 +85,7 @@ ClawRouter does **not** collect or forward third-party provider API keys. You do
 | ----------- | --------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
 | `walletKey` | Yes       | EVM private key used to sign USDC micropayments via x402. **Auto-generated locally on first run** — no user input required. Never transmitted over the network; only detached payment signatures are sent. |
 | `solanaKey` | Yes       | Solana keypair (BIP-44 `m/44'/501'/0'/0'`). Auto-derived from the same local mnemonic via `@scure/bip32` + `@scure/bip39`.                                                                                 |
-| `gateway`   | No        | Gateway URL. Defaults: `https://blockrun.ai/api` (Base) · `https://sol.blockrun.ai/api` (Solana).                                                                                                          |
+| `gateway`   | No        | Gateway URL. Defaults: `https://sol.blockrun.ai/api` (Solana, default chain for new installs) · `https://blockrun.ai/api` (Base, default for pre-existing installs).                                       |
 | `routing`   | No        | Optional override of the default four-tier router.                                                                                                                                                         |
 
 **How and where keys are stored:**
@@ -141,7 +141,7 @@ Rules handle ~~80% of requests in <1ms. Only ambiguous queries hit the LLM class
 
 ## Available Models
 
-<!-- br:models.chatVisible -->71<!-- /br:models.chatVisible --> models including: gpt-5.6-terra [balanced, stable default], gpt-5.6-sol [flagship], gpt-5.6-luna [cost-efficient], gpt-5.5, gpt-5.5-pro [max compute], chat-latest [ChatGPT Instant], gpt-5.4, gpt-4o, o3, claude-fable-5, claude-opus-5 [Anthropic flagship, 1M ctx], claude-opus-4.8, claude-opus-4.7, claude-opus-4.5, claude-sonnet-5, claude-sonnet-4.6, gemini-3.1-pro, gemini-2.5-flash, deepseek-v4-pro, deepseek-chat, grok-4.5, grok-4.3, grok-build-0.1, kimi-k3 [1M ctx flagship], kimi-k2.7, qwen3.7-max [Qwen flagship, 1M ctx], and 6 free NVIDIA models (deepseek-v4-flash [1M context], nemotron-3-nano-omni-30b-a3b-reasoning [vision], mistral-nemotron, step-3.7-flash, nemotron-nano-9b-v2, nemotron-nano-12b-v2-vl [vision]).
+<!-- br:models.chatVisible -->71<!-- /br:models.chatVisible --> models including: gpt-5.6-terra [balanced, stable default], gpt-5.6-sol [flagship], gpt-5.6-luna [cost-efficient], gpt-5.6-sol/terra/luna-pro [pro reasoning tiers], gpt-5.5, gpt-5.5-pro [max compute], chat-latest [ChatGPT Instant], gpt-5.4, gpt-4o, o3, claude-fable-5, claude-opus-5 [Anthropic flagship, 1M ctx], claude-opus-4.8, claude-opus-4.7, claude-opus-4.5, claude-sonnet-5, claude-sonnet-4.6, gemini-3.1-pro, gemini-3.6-flash, gemini-3.5-flash-lite, gemini-2.5-flash, deepseek-v4-pro, deepseek-chat, grok-4.5, grok-4.3, grok-build-0.1, kimi-k3 [1M ctx flagship], kimi-k2.7, qwen3.7-max [Qwen flagship, 1M ctx], qwen3.7-plus, qwen3.7-flash, hy3 [Tencent], mimo [Xiaomi MiMo-V2.5 Pro, 1M ctx], and 5 free NVIDIA models (nemotron-3-nano-omni-30b-a3b-reasoning [vision], mistral-nemotron, step-3.7-flash, nemotron-nano-9b-v2, nemotron-nano-12b-v2-vl [vision]).
 
 ## Built-in Agent Tools
 

package/src/top-models.json

@@ -38,7 +38,6 @@
   "deepseek/deepseek-v4-pro",
   "deepseek/deepseek-chat",
   "deepseek/deepseek-reasoner",
-  "free/deepseek-v4-flash",
   "free/nemotron-3-nano-omni-30b-a3b-reasoning",
   "free/mistral-nemotron",
   "free/step-3.7-flash",