Qwen3.5-9B bake-off results, model named Mortdecai

Bake-off: qwen3.5:9b base model, 147 cases:
  - 70.1% command match (2x qwen3:8b baseline)
  - 15.6% needed syntax fixes
  - 29.9% miss (mostly God/prayer — no persona training)
  - Avg 7.5s, median 5.7s (thinking tokens)

Model officially named Mortdecai.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

eval/results/bakeoff_1773963920.json

@@ -0,0 +1,17 @@
+{
+  "timestamp": 1773963920,
+  "ollama_url": "http://192.168.0.141:11434",
+  "note": "Partial run \u2014 147 of 1542 cases before manual kill",
+  "summary": [
+    {
+      "model": "qwen3.5:9b",
+      "n": 147,
+      "cmd_match_%": 70.1,
+      "syntax_fixes_%": 15.6,
+      "miss_%": 29.9,
+      "avg_latency_ms": 7457,
+      "median_latency_ms": 5660,
+      "note": "Base model, no fine-tuning. 2x better than qwen3:8b base (70% vs 34%)"
+    }
+  ]
+}