@blockrun/clawrouter 0.12.243 → 0.12.245

package/dist/index.js

@@ -92,7 +92,6 @@ var init_top_models = __esm({
       "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",
@@ -292,6 +291,14 @@ var init_models = __esm({
       "gpt-5.6-sol": "openai/gpt-5.6-sol",
       "gpt-5.6-terra": "openai/gpt-5.6-terra",
       "gpt-5.6-luna": "openai/gpt-5.6-luna",
+      // Pro reasoning tiers (2026-08-03). Explicit pins only — the generic
+      // shorthands above stay on standard Terra.
+      "gpt-5.6-sol-pro": "openai/gpt-5.6-sol-pro",
+      "gpt-5.6-terra-pro": "openai/gpt-5.6-terra-pro",
+      "gpt-5.6-luna-pro": "openai/gpt-5.6-luna-pro",
+      "sol-pro": "openai/gpt-5.6-sol-pro",
+      "terra-pro": "openai/gpt-5.6-terra-pro",
+      "luna-pro": "openai/gpt-5.6-luna-pro",
       "gpt-5.5": "openai/gpt-5.5",
       "gpt-5.5-pro": "openai/gpt-5.5-pro",
       // ChatGPT Instant. `chat-latest` is a rolling upstream id — pinning it means
@@ -343,6 +350,20 @@ var init_models = __esm({
       "qwen3.7-max": "qwen/qwen3.7-max",
       "qwen-3.7-max": "qwen/qwen3.7-max",
       "qwen3-7-max": "qwen/qwen3.7-max",
+      // Plus/Flash tiers (2026-08-03) — explicit pins, same rule as Max.
+      "qwen3.7-plus": "qwen/qwen3.7-plus",
+      "qwen-3.7-plus": "qwen/qwen3.7-plus",
+      "qwen3.7-flash": "qwen/qwen3.7-flash",
+      "qwen-3.7-flash": "qwen/qwen3.7-flash",
+      // Tencent + Xiaomi (2026-07-25) — each maker has exactly one model, so the
+      // bare maker names are safe to bind.
+      hy3: "tencent/hy3",
+      tencent: "tencent/hy3",
+      hunyuan: "tencent/hy3",
+      mimo: "xiaomi/mimo-v2.5-pro",
+      "mimo-v2.5-pro": "xiaomi/mimo-v2.5-pro",
+      "mimo-v2.5": "xiaomi/mimo-v2.5-pro",
+      xiaomi: "xiaomi/mimo-v2.5-pro",
       // Google
       // gemini-3-pro-preview delisted by Google 2026-06-06 — mirror the gateway
       // redirect to its successor so pinned callers land on 3.1-pro, not an error.
@@ -358,7 +379,10 @@ var init_models = __esm({
       flash: "google/gemini-2.5-flash",
       "gemini-3.1-pro-preview": "google/gemini-3.1-pro",
       "google/gemini-3.1-pro-preview": "google/gemini-3.1-pro",
+      "gemini-3.6-flash": "google/gemini-3.6-flash",
+      "gemini-3.6": "google/gemini-3.6-flash",
       "gemini-3.5-flash": "google/gemini-3.5-flash",
+      "gemini-3.5-flash-lite": "google/gemini-3.5-flash-lite",
       "gemini-3.1-flash-lite": "google/gemini-3.1-flash-lite",
       "gemini-2.5-flash-lite": "google/gemini-2.5-flash-lite",
       // xAI — grok-4.3 is the public flagship since 2026-06-04 (grok-3 and the
@@ -394,12 +418,17 @@ var init_models = __esm({
       "gpt-20b": "free/gpt-oss-20b",
       "nvidia/gpt-oss-120b": "free/gpt-oss-120b",
       "nvidia/gpt-oss-20b": "free/gpt-oss-20b",
-      // deepseek family → v4-flash, matching blockrun's 2026-07-17 redirect map
-      // (v4-flash recovered in the live re-probe; v3.2/v4-pro stay dead upstream).
-      "nvidia/deepseek-v3.2": "free/deepseek-v4-flash",
-      "free/deepseek-v3.2": "free/deepseek-v4-flash",
-      "nvidia/deepseek-v4-pro": "free/deepseek-v4-flash",
-      "free/deepseek-v4-pro": "free/deepseek-v4-flash",
+      // deepseek free family: v4-flash EOL'd 2026-08-12 (HTTP 410 from NVIDIA on both
+      // probe passes; prod gate fired [ALERT][free-model-dead] kind=gone twice that
+      // morning). The whole nvidia/deepseek-* family is now dead upstream — blockrun
+      // hid flash and retargeted v3.2/v4-pro (whose redirects chained through flash)
+      // straight to gpt-oss-120b. Ids naming v4-flash itself keep the real id (the
+      // gateway redirects them); the v3.2/v4-pro ids follow blockrun's retarget
+      // rather than chaining through a second dead model.
+      "nvidia/deepseek-v3.2": "free/gpt-oss-120b",
+      "free/deepseek-v3.2": "free/gpt-oss-120b",
+      "nvidia/deepseek-v4-pro": "free/gpt-oss-120b",
+      "free/deepseek-v4-pro": "free/gpt-oss-120b",
       "nvidia/deepseek-v4-flash": "free/deepseek-v4-flash",
       "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning": "free/nemotron-3-nano-omni-30b-a3b-reasoning",
       // qwen3-coder-480b retired (NVIDIA EOL 2026-06-14). Its server-side redirect used
@@ -441,13 +470,16 @@ var init_models = __esm({
       "mistral-large-3-675b": "free/mistral-large-3-675b",
       "qwen3.5-122b": "free/qwen3.5-122b-a10b",
       "qwen3-122b": "free/qwen3.5-122b-a10b",
-      // Free model shorthand aliases (v4-flash recovered 2026-07-17 — live again)
-      "deepseek-free": "free/deepseek-v4-flash",
-      "deepseek-v4-pro": "free/deepseek-v4-flash",
-      // free shorthand → flash (pro still hung)
+      // Free model shorthand aliases. v4-flash EOL'd 2026-08-12 — no free DeepSeek
+      // is left anywhere, so the generic "a free deepseek" shorthand follows the
+      // gateway's retarget (blockrun's /free-deepseek page points at gpt-oss-120b
+      // too). Shorthands naming v4-flash itself stay on the real id — the gateway
+      // redirects them, same treatment as seed-oss/mistral-large pins.
+      "deepseek-free": "free/gpt-oss-120b",
+      "deepseek-v4-pro": "free/gpt-oss-120b",
+      // free shorthand; pro dead upstream (410)
       "deepseek-v4-flash": "free/deepseek-v4-flash",
-      "v4-pro": "free/deepseek-v4-flash",
-      // V4 Pro NVIDIA hung → flash
+      "v4-pro": "free/gpt-oss-120b",
       "v4-flash": "free/deepseek-v4-flash",
       "mistral-free": "free/mistral-nemotron",
       // mistral-large-3-675b EOL'd 2026-07-28 → last live free Mistral (blockrun demoted its /free-mistral primary the same way)
@@ -527,16 +559,32 @@ var init_models = __esm({
       banana: "google/nano-banana",
       "banana-pro": "google/nano-banana-pro",
       "nano-banana-pro": "google/nano-banana-pro",
+      // Nano Banana 2 (Gemini 3.1 Flash imagegen, 2026-08-03). Explicit pins —
+      // bare `nano-banana`/`banana` stay on the original.
+      "nano-banana-2": "google/nano-banana-2",
+      "banana-2": "google/nano-banana-2",
       seedream: "bytedance/seedream-5-pro",
       "grok-imagine": "xai/grok-imagine-image",
       "grok-imagine-pro": "xai/grok-imagine-image-pro",
       cogview: "zai/cogview-4",
       // Video generation
       "grok-video": "xai/grok-imagine-video",
+      // Bare `seedance` deliberately stays on 1.5-pro. It is the cheapest of the
+      // family ($0.070/s vs 2.5's $0.315/s) and `/videogen` documents it as
+      // "default — cheapest"; repointing it at the newest tier would 4.5x the
+      // quote for every caller who typed the short name expecting the default.
+      // Same reasoning that keeps `kimi` on K2.7. Pin 2.5 explicitly to opt in.
       seedance: "bytedance/seedance-1.5-pro",
       "seedance-1.5": "bytedance/seedance-1.5-pro",
       "seedance-2-fast": "bytedance/seedance-2.0-fast",
-      "seedance-2": "bytedance/seedance-2.0"
+      "seedance-2": "bytedance/seedance-2.0",
+      "seedance-2.5": "bytedance/seedance-2.5",
+      "seedance-2-5": "bytedance/seedance-2.5",
+      // Seedance 2.0 Mini (2026-08-12): 720p + synced audio at half the flagship
+      // rate. Note it is NOT cheaper than 1.5-pro, so bare `seedance` stays put.
+      "seedance-2-mini": "bytedance/seedance-2.0-mini",
+      "seedance-2.0-mini": "bytedance/seedance-2.0-mini",
+      "seedance-mini": "bytedance/seedance-2.0-mini"
     };
     BLOCKRUN_MODELS = [
       // Smart routing meta-models — proxy replaces with actual model
@@ -644,11 +692,12 @@ var init_models = __esm({
         toolCalling: true
       },
       {
+        // Terra/Luna repriced 2026-07-30 (OpenAI price cut, blockrun #326).
         id: "openai/gpt-5.6-terra",
         name: "GPT-5.6 Terra",
         version: "5.6",
-        inputPrice: 2.5,
-        outputPrice: 15,
+        inputPrice: 2,
+        outputPrice: 12,
         contextWindow: 105e4,
         maxOutput: 128e3,
         reasoning: true,
@@ -660,10 +709,52 @@ var init_models = __esm({
         id: "openai/gpt-5.6-luna",
         name: "GPT-5.6 Luna",
         version: "5.6",
+        inputPrice: 0.2,
+        outputPrice: 1.2,
+        contextWindow: 105e4,
+        maxOutput: 128e3,
+        vision: true,
+        agentic: true,
+        toolCalling: true
+      },
+      // GPT-5.6 Pro tiers (blockrun #329, 2026-08-03): each base tier with pro
+      // reasoning mode. Terra Pro lands at half the standard Terra rate; Luna Pro
+      // is the budget deep-reasoning tier.
+      {
+        id: "openai/gpt-5.6-sol-pro",
+        name: "GPT-5.6 Sol Pro",
+        version: "5.6",
+        inputPrice: 5,
+        outputPrice: 30,
+        contextWindow: 105e4,
+        maxOutput: 128e3,
+        reasoning: true,
+        vision: true,
+        agentic: true,
+        toolCalling: true
+      },
+      {
+        id: "openai/gpt-5.6-terra-pro",
+        name: "GPT-5.6 Terra Pro",
+        version: "5.6",
         inputPrice: 1,
         outputPrice: 6,
         contextWindow: 105e4,
         maxOutput: 128e3,
+        reasoning: true,
+        vision: true,
+        agentic: true,
+        toolCalling: true
+      },
+      {
+        id: "openai/gpt-5.6-luna-pro",
+        name: "GPT-5.6 Luna Pro",
+        version: "5.6",
+        inputPrice: 0.1,
+        outputPrice: 0.6,
+        contextWindow: 105e4,
+        maxOutput: 128e3,
+        reasoning: true,
         vision: true,
         agentic: true,
         toolCalling: true
@@ -1067,19 +1158,45 @@ var init_models = __esm({
         toolCalling: true
       },
       {
-        // Latest-generation Flash with built-in thinking mode (backend
-        // 2026-05-19) — frontier-class quality at Flash pricing.
+        // Newest-generation Flash with built-in thinking mode (blockrun #329,
+        // 2026-08-03). 17% cheaper output than 3.5 Flash.
+        id: "google/gemini-3.6-flash",
+        name: "Gemini 3.6 Flash",
+        version: "3.6",
+        inputPrice: 1.5,
+        outputPrice: 7.5,
+        contextWindow: 1048576,
+        maxOutput: 65536,
+        reasoning: true,
+        vision: true,
+        toolCalling: true
+      },
+      {
+        // Repriced 0.5/3.0 → 1.5/9.0 (blockrun #304: it was billed at 1/3 of
+        // Google's real rate).
         id: "google/gemini-3.5-flash",
         name: "Gemini 3.5 Flash",
         version: "3.5",
-        inputPrice: 0.5,
-        outputPrice: 3,
+        inputPrice: 1.5,
+        outputPrice: 9,
         contextWindow: 1048576,
         maxOutput: 65536,
         reasoning: true,
         vision: true,
         toolCalling: true
       },
+      {
+        // Ultra-fast lightweight tier with thinking mode (blockrun #329).
+        id: "google/gemini-3.5-flash-lite",
+        name: "Gemini 3.5 Flash Lite",
+        version: "3.5",
+        inputPrice: 0.3,
+        outputPrice: 2.5,
+        contextWindow: 1048576,
+        maxOutput: 65536,
+        reasoning: true,
+        toolCalling: true
+      },
       {
         id: "google/gemini-3-flash-preview",
         name: "Gemini 3 Flash Preview",
@@ -1136,12 +1253,14 @@ var init_models = __esm({
       },
       // DeepSeek — V4 family (2026-04-24). The legacy deepseek-chat/reasoner
       // aliases are served upstream as V4 Flash non-thinking / thinking modes.
+      // Repriced 0.20/0.40 → 0.14/0.28 (blockrun #354: DeepSeek's published
+      // deepseek-v4-flash rate; the old numbers were 1.43x the real rate).
       {
         id: "deepseek/deepseek-chat",
         name: "DeepSeek V4 Flash Chat",
         version: "4-flash",
-        inputPrice: 0.2,
-        outputPrice: 0.4,
+        inputPrice: 0.14,
+        outputPrice: 0.28,
         contextWindow: 1e6,
         maxOutput: 8192,
         toolCalling: true
@@ -1150,8 +1269,8 @@ var init_models = __esm({
         id: "deepseek/deepseek-reasoner",
         name: "DeepSeek V4 Flash Reasoner",
         version: "4-flash",
-        inputPrice: 0.2,
-        outputPrice: 0.4,
+        inputPrice: 0.14,
+        outputPrice: 0.28,
         contextWindow: 1e6,
         maxOutput: 8192,
         reasoning: true,
@@ -1230,6 +1349,32 @@ var init_models = __esm({
         agentic: true,
         toolCalling: true
       },
+      // Qwen3.7 Plus/Flash (blockrun #329, 2026-08-03): the balanced and
+      // latency tiers under Max. Plus genuinely caps output at 131072 while
+      // Flash and Max cap at 65536 (endpoint-probed upstream).
+      {
+        id: "qwen/qwen3.7-plus",
+        name: "Qwen3.7 Plus",
+        version: "3.7-plus",
+        inputPrice: 0.32,
+        outputPrice: 1.28,
+        contextWindow: 1e6,
+        maxOutput: 131072,
+        reasoning: true,
+        agentic: true,
+        toolCalling: true
+      },
+      {
+        id: "qwen/qwen3.7-flash",
+        name: "Qwen3.7 Flash",
+        version: "3.7-flash",
+        inputPrice: 0.03,
+        outputPrice: 0.13,
+        contextWindow: 1e6,
+        maxOutput: 65536,
+        reasoning: true,
+        toolCalling: true
+      },
       // Kimi K2.6 — superseded by K2.7 (2026-06-13), hidden on BlockRun but still routable.
       {
         id: "moonshot/kimi-k2.6",
@@ -1441,6 +1586,34 @@ var init_models = __esm({
         agentic: true,
         toolCalling: true
       },
+      // Tencent + Xiaomi (blockrun 2026-07-25): the two largest real demand gaps
+      // in the catalog — Hy3 held the #1 usage slot on the biggest public
+      // aggregator for 19 days; MiMo ran ~22% weekly share in April. Both are
+      // reasoning models resold via blockrun's OpenRouter pool. toolCalling
+      // LIVE-VERIFIED 2026-08-12: both returned a structured tool_calls array
+      // (name + valid JSON args, finish_reason "tool_calls") through the gateway.
+      {
+        id: "tencent/hy3",
+        name: "Tencent Hy3",
+        version: "hy3",
+        inputPrice: 0.132,
+        outputPrice: 0.528,
+        contextWindow: 262144,
+        maxOutput: 128e3,
+        reasoning: true,
+        toolCalling: true
+      },
+      {
+        id: "xiaomi/mimo-v2.5-pro",
+        name: "Xiaomi MiMo-V2.5 Pro",
+        version: "v2.5-pro",
+        inputPrice: 0.435,
+        outputPrice: 0.87,
+        contextWindow: 1048576,
+        maxOutput: 131072,
+        reasoning: true,
+        toolCalling: true
+      },
       // MiniMax
       {
         id: "minimax/minimax-m3",
@@ -1514,11 +1687,13 @@ var init_models = __esm({
         maxOutput: 16384
       },
       {
-        // V4 Flash: 284B / 13B active MoE, 1M context. ~5x faster than V4 Pro.
-        // Strong on chat/summarization (MMLU-Pro 86.2). Caveat: weaker factual
-        // recall (SimpleQA 34% vs Pro's 58%) — pick V4 Pro for fact-heavy loops.
-        // Recovered in blockrun's 2026-07-17 live re-probe (3.2s both passes) —
-        // un-hidden upstream and back in the advertised free 8.
+        // V4 Flash: 284B / 13B active MoE, 1M context. EOL'd 2026-08-12 — NVIDIA
+        // published 410 Gone ("has reached its end of life") on both probe passes
+        // and blockrun's prod gate fired [ALERT][free-model-dead] kind=gone twice
+        // that morning. The whole nvidia/deepseek-* family is dead upstream; there
+        // is no free DeepSeek left anywhere. Blockrun hid it and redirects calls to
+        // gpt-oss-120b. Entry kept so explicit pins stay routable; off the picker,
+        // the FREE_MODELS cascade, and the router fallback chains.
         id: "free/deepseek-v4-flash",
         name: "[Free] DeepSeek V4 Flash",
         version: "v4-flash",
@@ -1602,7 +1777,8 @@ var init_models = __esm({
       // 2026-06-16: BlockRun's 2026-06-14 free-tier refresh (self-healing health gate
       // + probe-verified lineup, blockrun commit 5817ecd) added these live NVIDIA free
       // models. Status per blockrun's 2026-07-17 live re-probe: qwen3-coder-480b and
-      // glm-4.7 stay dead (server-redirected); deepseek-v4-flash recovered (see above).
+      // glm-4.7 stay dead (server-redirected); deepseek-v4-flash recovered then
+      // EOL'd for good 2026-08-12 (see above).
       {
         // Qwen3-Next 80B (A3B active MoE): 262K context. DIED in the 2026-07-17
         // re-probe (">60s / DEGRADED") — hidden upstream, gateway redirects pinned
@@ -1707,11 +1883,12 @@ var init_models = __esm({
       },
       {
         // Flat-rate launch promo ended 2026-06-06 — backend bills per-token now.
+        // Repriced 0.60/1.92 → 1.00/3.20 (blockrun #354 correction).
         id: "zai/glm-5",
         name: "GLM-5",
         version: "5",
-        inputPrice: 0.6,
-        outputPrice: 1.92,
+        inputPrice: 1,
+        outputPrice: 3.2,
         contextWindow: 2e5,
         maxOutput: 128e3,
         reasoning: true,
@@ -37866,10 +38043,9 @@ ${value.slice(-(scanLimit - prefixLength))}`;
           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.
+            // deepseek-v4-flash and seed-oss-36b sat here until NVIDIA EOL'd them
+            // (410; 2026-08-12 and 2026-08-03 respectively). gpt-oss-120b/20b already
+            // head this chain, so the rungs are dropped, not retargeted.
             "google/gemini-3.1-flash-lite",
             // $0.25/$1.50 — newest flash-lite
             "openai/gpt-5.4-nano",
@@ -92400,8 +92576,6 @@ var init_proxy = __esm({
     FREE_MODELS = /* @__PURE__ */ new Set([
       "free/gpt-oss-120b",
       "free/gpt-oss-20b",
-      "free/deepseek-v4-flash",
-      // 1M ctx; slow (~10 tok/s, 07-28 probe) but completes
       "free/mistral-nemotron",
       // strong instruction following
       "free/step-3.7-flash",
@@ -92498,6 +92672,8 @@ var init_proxy = __esm({
         sizes: { "1024x1024": 0.045, "2048x1024": 0.045, "2048x2048": 0.09 }
       },
       "google/nano-banana": { default: 0.05 },
+      // Nano Banana 2 (Gemini 3.1 Flash imagegen, blockrun #329): $0.09/image at 1K.
+      "google/nano-banana-2": { default: 0.09, sizes: { "1024x1024": 0.09 } },
       "google/nano-banana-pro": {
         default: 0.1,
         sizes: { "1024x1024": 0.1, "2048x2048": 0.1, "4096x4096": 0.15 }
@@ -92518,13 +92694,32 @@ var init_proxy = __esm({
     };
     VIDEO_PRICING = {
       "xai/grok-imagine-video": { pricePerSecond: 0.05, defaultDurationSeconds: 8 },
-      "bytedance/seedance-1.5-pro": { pricePerSecond: 0.0875, defaultDurationSeconds: 5 },
+      // Seedance rates re-synced 2026-08-08 from blockrun's models.ts after the
+      // family reprice (blockrun #351/#354). All three were carrying pre-reprice
+      // numbers, which only ever skewed the usage log — the charge comes from the
+      // x402 header, and estimateVideoCost is the fallback for when it is absent.
+      "bytedance/seedance-1.5-pro": { pricePerSecond: 0.07, defaultDurationSeconds: 5 },
       "bytedance/seedance-2.0-fast": {
-        pricePerSecond: 0.22687,
+        pricePerSecond: 0.165,
         defaultDurationSeconds: 5
       },
       "bytedance/seedance-2.0": {
-        pricePerSecond: 0.28358,
+        pricePerSecond: 0.227,
+        defaultDurationSeconds: 5
+      },
+      // Seedance 2.0 Mini (blockrun #366/#368, 2026-08-12): 720p + synced audio at
+      // half the flagship rate, 480p/720p only, 4–15s. blockrun signs the 3dp-floored
+      // per-second value (0.079, not the raw 0.0797) — mirror the signed number.
+      "bytedance/seedance-2.0-mini": {
+        pricePerSecond: 0.079,
+        defaultDurationSeconds: 5
+      },
+      // Seedance 2.5: 720p with synced audio, up to 30s. Not a replacement for 2.0
+      // Pro — that is still the one for 1080p/4K. No OpenRouter failover exists for
+      // 2.5, so a token360 outage surfaces as an error instead of quietly rendering
+      // on another model.
+      "bytedance/seedance-2.5": {
+        pricePerSecond: 0.315,
         defaultDurationSeconds: 5
       },
       // Sora 2 via Azure AI Foundry — flat $0.10/s for both t2v and i2v.
@@ -225799,7 +225994,9 @@ function buildVideoGenerationProvider() {
       "xai/grok-imagine-video",
       "bytedance/seedance-1.5-pro",
       "bytedance/seedance-2.0-fast",
-      "bytedance/seedance-2.0"
+      "bytedance/seedance-2.0",
+      "bytedance/seedance-2.0-mini",
+      "bytedance/seedance-2.5"
     ],
     capabilities: {
       maxVideos: 1,
@@ -226299,7 +226496,7 @@ var init_index = __esm({
             const parsed = parseGenArgs(ctx.args ?? "");
             if (!parsed.prompt) {
               return {
-                text: "Usage: `/cr-imagegen <prompt> [--model=<alias>] [--size=1024x1024] [--n=1]`\n\nAliases: `nano-banana` (default), `banana-pro`, `dalle`, `gpt-image`, `flux`, `grok-imagine`, `grok-imagine-pro`, `cogview`."
+                text: "Usage: `/cr-imagegen <prompt> [--model=<alias>] [--size=1024x1024] [--n=1]`\n\nAliases: `nano-banana` (default), `banana-2` (Gemini 3.1 Flash quality), `banana-pro`, `dalle`, `gpt-image`, `flux`, `grok-imagine`, `grok-imagine-pro`, `cogview`."
               };
             }
             const model = resolveModelAlias(parsed.model ?? "nano-banana");
@@ -226351,7 +226548,7 @@ ${errText}`
             const parsed = parseGenArgs(ctx.args ?? "");
             if (!parsed.prompt) {
               return {
-                text: "Usage: `/videogen <prompt> [--model=<alias>] [--duration=5|8|10]`\n\nAliases: `seedance` (1.5-pro, default \u2014 cheapest), `seedance-2-fast`, `seedance-2`, `grok-video`.\n\n\u23F1\uFE0F  Generation takes 60\u2013180 seconds. Payment settles only on success."
+                text: "Usage: `/videogen <prompt> [--model=<alias>] [--duration=5|8|10]`\n\nAliases: `seedance` (1.5-pro, default \u2014 cheapest), `seedance-2-fast`, `seedance-2`, `seedance-2-mini` (720p + audio, half the flagship rate), `seedance-2.5` (720p + synced audio, long-form), `grok-video`.\n\n\u23F1\uFE0F  Generation takes 60\u2013180 seconds. Payment settles only on success."
               };
             }
             const model = resolveModelAlias(parsed.model ?? "seedance");

package/dist/router/index.js

@@ -3008,10 +3008,9 @@ var DEFAULT_ROUTING_CONFIG = {
       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.
+        // deepseek-v4-flash and seed-oss-36b sat here until NVIDIA EOL'd them
+        // (410; 2026-08-12 and 2026-08-03 respectively). gpt-oss-120b/20b already
+        // head this chain, so the rungs are dropped, not retargeted.
         "google/gemini-3.1-flash-lite",
         // $0.25/$1.50 — newest flash-lite
         "openai/gpt-5.4-nano",

package/docs/anthropic-cost-savings.md

@@ -50,7 +50,7 @@ This is where you're paying for real value:
 
 ## The Solution: ClawRouter
 
-[ClawRouter](https://github.com/BlockRunAI/ClawRouter) is an open-source local proxy that sits between your app and <!-- br:models.chatVisible -->71<!-- /br:models.chatVisible --> AI models. It saves you money in three ways: **smart routing**, **token optimization**, and **response caching**.
+[ClawRouter](https://github.com/BlockRunAI/ClawRouter) is an open-source local proxy that sits between your app and <!-- br:models.chatVisible -->70<!-- /br:models.chatVisible --> AI models. It saves you money in three ways: **smart routing**, **token optimization**, and **response caching**.
 
 ```
 ┌─────────────┐     ┌──────────────────────────────┐     ┌──────────────────┐

package/docs/anthropic-third-party-harness-changes.md

@@ -31,7 +31,7 @@ This is exactly the problem [ClawRouter](https://github.com/BlockRunAI/ClawRoute
 
 ## ClawRouter: Smart Routing for Agents
 
-ClawRouter is an open-source local proxy that sits between your agent and <!-- br:models.chatVisible -->71<!-- /br:models.chatVisible --> LLM models across 9 providers. It analyzes every request across 15 dimensions and routes it to the cheapest model that can handle it — in under 1ms, entirely locally.
+ClawRouter is an open-source local proxy that sits between your agent and <!-- br:models.chatVisible -->70<!-- /br:models.chatVisible --> LLM models across 9 providers. It analyzes every request across 15 dimensions and routes it to the cheapest model that can handle it — in under 1ms, entirely locally.
 
 ```
 Your Agent → ClawRouter (localhost:8402) → Best model for the job
@@ -56,7 +56,7 @@ From 20,000+ production requests:
 | Free models (trivial tasks)      | 12.8%        | $0.00             |
 | Others                           | 13.8%        | varies            |
 
-**Result: <!-- br:savings.autoVsBaselinePct -->87<!-- /br:savings.autoVsBaselinePct -->% cheaper than pinning Claude Opus 5 for every request** — the published figure, priced on a stated workload mix rather than estimated.
+**Result: <!-- br:savings.autoVsBaselinePct -->88<!-- /br:savings.autoVsBaselinePct -->% cheaper than pinning Claude Opus 5 for every request** — the published figure, priced on a stated workload mix rather than estimated.
 
 A typical user running 10K mixed requests/month:
 
@@ -74,7 +74,7 @@ Then point your agent at `http://localhost:8402/v1/` with any OpenAI-compatible
 
 - **No API keys to manage** — wallet-based cryptographic auth
 - **No subscriptions** — pay per request in USDC (Base or Solana)
-- **No vendor lock-in** — <!-- br:models.chatVisible -->71<!-- /br:models.chatVisible --> models, switch anytime
+- **No vendor lock-in** — <!-- br:models.chatVisible -->70<!-- /br:models.chatVisible --> models, switch anytime
 - **You control your wallet** — non-custodial, funds never held by a third party
 - **Budget caps** — set a max spend per session, ClawRouter gracefully downgrades when budget runs low
 - **Token compression** — 7-layer pipeline reduces token costs by 7–40% before they hit any provider

package/docs/clawrouter-cuts-llm-api-costs-500x.md

@@ -135,7 +135,7 @@ Less context repeated = fewer tokens = lower cost. Agents that need to recall ea
 
 ### 7. x402 Micropayments — Wallet-Based Budget Control
 
-<p align="center"><img src="assets/blockrun-clawrouter-x402-usdc-micropayment-wallet-budget-control.png" alt="Budget limits enforced by physical construction — wallet loaded via Base/Solana, pay per call across <!-- br:models.chatVisible -->71<!-- /br:models.chatVisible --> models, balance hits zero and the valve shuts cleanly" width="720"></p>
+<p align="center"><img src="assets/blockrun-clawrouter-x402-usdc-micropayment-wallet-budget-control.png" alt="Budget limits enforced by physical construction — wallet loaded via Base/Solana, pay per call across <!-- br:models.chatVisible -->70<!-- /br:models.chatVisible --> models, balance hits zero and the valve shuts cleanly" width="720"></p>
 
 ClawRouter pays for inference via [x402](https://x402.org/) USDC micropayments (Base or Solana). You load a wallet. Each inference call costs exactly what it costs. When the wallet runs low, requests stop cleanly.
 

package/docs/clawrouter-vs-openrouter-llm-routing-comparison.md

@@ -222,7 +222,7 @@ OpenRouter doesn't always pass through provider-specific features correctly. Ima
 
 **Direct provider routing.** ClawRouter routes through BlockRun's API directly to providers — not through a second aggregator. One hop, not two. Provider-specific features work because there's no middleman translating them.
 
-![Guaranteed Feature Parity & Direct Connectivity — Three-panel diagram: Vision (image_url auto-detected → vision-capable models only), Tool Calling (toolCalling flag → agentic models only), Catalog (curated <!-- br:models.chatVisible -->71<!-- /br:models.chatVisible --> models with automatic legacy-to-modern redirects). Direct provider routing means no dropped payloads.](./assets/clawrouter-feature-parity-direct-connectivity.png)
+![Guaranteed Feature Parity & Direct Connectivity — Three-panel diagram: Vision (image_url auto-detected → vision-capable models only), Tool Calling (toolCalling flag → agentic models only), Catalog (curated <!-- br:models.chatVisible -->70<!-- /br:models.chatVisible --> models with automatic legacy-to-modern redirects). Direct provider routing means no dropped payloads.](./assets/clawrouter-feature-parity-direct-connectivity.png)
 
 ---
 
@@ -240,7 +240,7 @@ When new models launch, OpenRouter's catalog lags. Users configure a model that
 
 ### How ClawRouter Solves This
 
-ClawRouter maintains a curated catalog of <!-- br:models.chatVisible -->71<!-- /br:models.chatVisible --> models across 9 providers (including 9 free models), updated with each release. Delisted models have automatic redirect aliases:
+ClawRouter maintains a curated catalog of <!-- br:models.chatVisible -->70<!-- /br:models.chatVisible --> models across 9 providers (including 9 free models), updated with each release. Delisted models have automatic redirect aliases:
 
 ```typescript
 // Delisted models redirect automatically
@@ -268,10 +268,10 @@ No silent drops. No stale catalog. Models are benchmarked for speed, quality, an
 | **Rate limits**     | Per-key, shared                  | Per-wallet, independent                                                                                  |
 | **Vision support**  | Images sometimes dropped         | Auto-detected, vision-only fallback                                                                      |
 | **Tool calling**    | Silent failures with some models | Flag-based filtering, guaranteed support                                                                 |
-| **Model catalog**   | Laggy, silent drops              | Curated <!-- br:models.chatVisible -->71<!-- /br:models.chatVisible --> models, redirect aliases         |
+| **Model catalog**   | Laggy, silent drops              | Curated <!-- br:models.chatVisible -->70<!-- /br:models.chatVisible --> models, redirect aliases         |
 | **Budget control**  | Monthly invoice                  | Per-session cap (`maxCostPerRun`)                                                                        |
 | **Setup**           | Create account, paste key        | Agent generates wallet, auto-configured                                                                  |
-| **Average cost**    | $25/M tokens (Opus direct)       | auto-routed = **<!-- br:savings.autoVsBaselinePct -->87<!-- /br:savings.autoVsBaselinePct -->% savings** |
+| **Average cost**    | $25/M tokens (Opus direct)       | auto-routed = **<!-- br:savings.autoVsBaselinePct -->88<!-- /br:savings.autoVsBaselinePct -->% savings** |
 
 ![The Engineering Matrix — Side-by-side feature comparison: OpenRouter vs ClawRouter across Routing, Authentication, Payment, Fallback, Model IDs, Empty Wallet, Vision/Tools, and Average Cost. ClawRouter wins on every dimension.](./assets/clawrouter-engineering-matrix-comparison.png)
 

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@blockrun/clawrouter",
-  "version": "0.12.243",
-  "description": "Smart LLM router — save 87% on inference costs. 71 models (6 free), one wallet, x402 micropayments.",
+  "version": "0.12.245",
+  "description": "Smart LLM router — save 88% on inference costs. 70 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/18bf4abc7e34599e51595b90703c385f53d4f4a6",
     "@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 87% 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 70 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 -->87<!-- /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 -->70<!-- /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.
 
@@ -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 -->70<!-- /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