chore: sync capabilities: add google/gemini-4-argon (#9)

Regenerate model-capabilities.ts with scripts/sync-model-capabilities.mjs
from GET https://blockrun.ai/api/v1/models (90 text models). Gemini 4 Argon
(live on Base and Solana since 2026-10-20) is added with a 1M context window,
65,536 output cap and vision. supportsTools is false: there is no paid
tool-call probe for it in probe-tools.json yet, so it fails closed, the same
admission rule as GPT-6.1 Sol, Mistral Large 4 and DeepSeek V4.1 Flash.

Catalog-only by product decision: config.ts chains and portfolio are
untouched, and decisions.snapshot.json is byte-identical, so no routing
decision changed. README count updated; dist/ rebuilt.

Co-authored-by: 1bcMax <viewitter@gmail.com>

Author: VickyXAI

README.md

@@ -263,7 +263,7 @@ Two rules govern this file, and both are load-bearing:
 1. **These are weak priors, never task-quality labels.** They inform the `speed` and `reliability` terms only. A model is not "better" because it is fast.
 2. **Historical numbers are never presented as a current provider SLA.** Hosts are expected to inject fresher observations (see below); the committed snapshot exists so the engine is safe and useful when a catalog is temporarily unavailable.
 
-The repository ships 66 live performance profiles (2026-08-29, three samples per model), 9 auditable historical seeds, 74 function-calling probe results, and a built-in snapshot of all 89 text models in the public catalog (2026-10-20 sync). Eleven newly listed models passed a real structured tool-call probe and are admitted as availability fallbacks for tool-required turns. DeepSeek V4 Flash Vision Experimental and Qwen 3.8 Flash did not return a valid tool call and remain fail-closed. GPT-6.1 Sol, Mistral Large 4 and DeepSeek V4.1 Flash, listed since the last probe, have no probe evidence yet. A model with no probe evidence can serve plain chat but never a tool-required turn.
+The repository ships 66 live performance profiles (2026-08-29, three samples per model), 9 auditable historical seeds, 74 function-calling probe results, and a built-in snapshot of all 90 text models in the public catalog (2026-10-20 sync). Eleven newly listed models passed a real structured tool-call probe and are admitted as availability fallbacks for tool-required turns. DeepSeek V4 Flash Vision Experimental and Qwen 3.8 Flash did not return a valid tool call and remain fail-closed. GPT-6.1 Sol, Mistral Large 4, DeepSeek V4.1 Flash and Gemini 4 Argon, listed since the last probe, have no probe evidence yet. A model with no probe evidence can serve plain chat but never a tool-required turn.
 
 Passing the function-call probe does **not** promote a model to primary. New tool-capable models are appended after the configured and trajectory-benchmarked candidate chains with neutral task affinity; they can recover an unavailable route immediately (cheapest first outside Premium), while primary promotion still requires an end-to-end agent evaluation.
 

dist/index.js

@@ -187,6 +187,12 @@ export const DEFAULT_MODEL_CAPABILITIES: Readonly<Record<string, ModelCapabiliti
       supportsTools: true,
       supportsVision: true,
     },
+    "google/gemini-4-argon": { // supportsTools: not probed — fails closed
+      contextWindow: 1_048_576,
+      maxOutputTokens: 65_536,
+      supportsTools: false,
+      supportsVision: true,
+    },
     "minimax/minimax-m2.7": {
       contextWindow: 204_800,
       maxOutputTokens: 16_384,