VLM migration: qwen3.5-9b-vlm → gemma-4-12b across entire harness

- router/router.py: model ID, context window (131K→262K), VRAM comment
- litellm_config.yaml: model name updated
- gpu-router.conf, gpu-router-docker.conf: pool names, comments
- dashboard/dashboard.py: MODELS array refactor for single-point config
- README.md: architecture diagram, model table

router/router.py

@@ -10,31 +10,31 @@ GPU_LIGHT_URL = os.environ.get("GPU_LIGHT_URL", "http://192.168.68.110:8080/v1")
 GPU_SIDECARS = {
     "qwen3.6-35B-A3B": "http://192.168.68.15:8090",
     "qwen3.6-27B-code": "http://192.168.68.8:8090",
-    "qwen3.5-9b-vlm": "http://192.168.68.110:8090",
+    "gemma-4-12b": "http://192.168.68.110:8090",
 }
 GPU_URLS = {
     "qwen3.6-35B-A3B": GPU_MOE_URL,
     "qwen3.6-27B-code": GPU_DENSE_URL,
-    "qwen3.5-9b-vlm": GPU_LIGHT_URL,
+    "gemma-4-12b": GPU_LIGHT_URL,
 }
 # Max concurrent requests per GPU (based on llama.cpp --parallel)
 GPU_MAX_CONCURRENT = {
     "qwen3.6-35B-A3B": 2,   # 2 slots
     "qwen3.6-27B-code": 2,  # 2 slots
-    "qwen3.5-9b-vlm": 2,       # 2 slots (12GB VRAM, 4GB headroom)
+    "gemma-4-12b": 2,       # 2 slots (7.1GB VRAM)
 }
 
 # Context window sizes (tokens) — used for compaction signals
 GPU_CONTEXT = {
     "qwen3.6-35B-A3B": 131072,
     "qwen3.6-27B-code": 98304,
-    "qwen3.5-9b-vlm": 131072,
+    "gemma-4-12b": 262144,  # 262K context (gemma-4-12b)
 }
 
 TIER_MODELS = {
-    "starter": ["qwen3.5-9b-vlm"],
-    "professional": ["qwen3.6-35B-A3B", "qwen3.6-27B-code", "qwen3.5-9b-vlm"],
-    "enterprise": ["qwen3.6-35B-A3B", "qwen3.6-27B-code", "qwen3.5-9b-vlm"],
+    "starter": ["gemma-4-12b"],
+    "professional": ["qwen3.6-35B-A3B", "qwen3.6-27B-code", "gemma-4-12b"],
+    "enterprise": ["qwen3.6-35B-A3B", "qwen3.6-27B-code", "gemma-4-12b"],
 }
 API_KEYS = {
     "sk-syslog-local-master-key": {"tier": "enterprise", "agent": "admin"},
@@ -149,7 +149,7 @@ def route(rd, tier):
     sys = any(m.get("role")=="system" for m in msgs)
     turns = len([m for m in msgs if m.get("role") in ("user","assistant")])
     hints = rd.get("routing_hints",{})
-    allowed = TIER_MODELS.get(tier, ["qwen3.5-9b-vlm"])
+    allowed = TIER_MODELS.get(tier, ["gemma-4-12b"])
     avail = [m for m in available_models() if m in allowed]
     if not avail: return {"model": allowed[0], "reason": "all_saturated", "saturated": True}
     # Check if all available GPUs are at max capacity
@@ -168,8 +168,8 @@ def route(rd, tier):
         return {"model": target, "reason": "explicit"}
     
     if hints:
-        if hints.get("priority")=="speed" and "qwen3.5-9b-vlm" in avail:
-            return select_best_gpu(["qwen3.5-9b-vlm"], "hint_speed") or {"model":"qwen3.5-9b-vlm","reason":"hint_speed"}
+        if hints.get("priority")=="speed" and "gemma-4-12b" in avail:
+            return select_best_gpu(["gemma-4-12b"], "hint_speed") or {"model":"gemma-4-12b","reason":"hint_speed"}
         if hints.get("priority")=="quality" and "qwen3.6-27B-code" in avail:
             return select_best_gpu(["qwen3.6-27B-code"], "hint_quality") or {"model":"qwen3.6-27B-code","reason":"hint_quality"}
     
@@ -177,18 +177,18 @@ def route(rd, tier):
     words = len(first_msg.split()) if isinstance(first_msg, str) else 99
     
     # TIER 1: Lightweight — single-turn short queries → VLM first
-    if not sys and turns <= 1 and words <= 100 and "qwen3.5-9b-vlm" in avail:
-        if not is_gpu_busy("qwen3.5-9b-vlm"):
-            return {"model":"qwen3.5-9b-vlm","reason":"lightweight"}
+    if not sys and turns <= 1 and words <= 100 and "gemma-4-12b" in avail:
+        if not is_gpu_busy("gemma-4-12b"):
+            return {"model":"gemma-4-12b","reason":"lightweight"}
         # VLM busy — fall back to Dense, then MoE
         fallback = [m for m in ["qwen3.6-35B-A3B","qwen3.6-27B-code"] if m in avail]
         result = select_best_gpu(fallback, "lightweight_fallback")
         if result: return result
     
     # TIER 2: Simple conversations — short context, any prompt → VLM preferred
-    if t <= 1000 and turns <= 4 and "qwen3.5-9b-vlm" in avail:
-        if not is_gpu_busy("qwen3.5-9b-vlm"):
-            return {"model":"qwen3.5-9b-vlm","reason":"simple_conv"}
+    if t <= 1000 and turns <= 4 and "gemma-4-12b" in avail:
+        if not is_gpu_busy("gemma-4-12b"):
+            return {"model":"gemma-4-12b","reason":"simple_conv"}
         # VLM busy — try Dense
         if "qwen3.6-27B-code" in avail and not is_gpu_busy("qwen3.6-27B-code"):
             return {"model":"qwen3.6-27B-code","reason":"simple_conv_fallback"}
@@ -196,13 +196,13 @@ def route(rd, tier):
     # TIER 3: Heavy reasoning — extremely large context or very long conversations
     if t > 50000 or turns > 25:
         # MoE first (131K context handles heavy sessions), then Dense (98K reasoning), then Light (131K fallback)
-        candidates = [m for m in ["qwen3.6-35B-A3B","qwen3.6-27B-code","qwen3.5-9b-vlm"] if m in avail]
+        candidates = [m for m in ["qwen3.6-35B-A3B","qwen3.6-27B-code","gemma-4-12b"] if m in avail]
         result = select_best_gpu(candidates, "heavy_reasoning")
         if result: return result
     
     # TIER 4: Default — MoE first, VLM helps, Dense last (slow)
     if t <= 50000:
-        candidates = [m for m in ["qwen3.6-35B-A3B","qwen3.5-9b-vlm","qwen3.6-27B-code"] if m in avail]
+        candidates = [m for m in ["qwen3.6-35B-A3B","gemma-4-12b","qwen3.6-27B-code"] if m in avail]
         result = select_best_gpu(candidates, "default")
         if result: return result