add gemma 3n, smallthinker 4b0.6b and falcon h1-1.5b

app.py

@@ -24,6 +24,28 @@ cancel_event = threading.Event()
 # Torch-Compatible Model Definitions with Adjusted Descriptions
 # ------------------------------
 MODELS = {
+    # … your existing entries …
+
+    # Gemma 3n “effective” variants (official Google repos)
+    "Gemma-3n-E2B": {
+        "repo_id":   "google/gemma-3n-E2B",
+        "description":"Gemma 3n base model with effective 2 B parameters (≈2 GB VRAM)"
+    },
+    "Gemma-3n-E4B": {
+        "repo_id":   "google/gemma-3n-E4B",
+        "description":"Gemma 3n base model with effective 4 B parameters (≈3 GB VRAM)"
+    },
+
+    # PowerInfer SmallThinker (instruction‑tuned)
+    "SmallThinker-4BA0.6B-Instruct": {
+        "repo_id":   "PowerInfer/SmallThinker-4BA0.6B-Instruct",
+        "description":"SmallThinker 4 B backbone with 0.6 B activated parameters, instruction‑tuned"
+    },
+    # TIIUAE Falcon‑H1 (instruction‑tuned)
+    "Falcon-H1-1.5B-Instruct": {
+        "repo_id":   "tiiuae/Falcon-H1-1.5B-Instruct",
+        "description":"Falcon‑H1 model with 1.5 B parameters, instruction‑tuned"
+    },
     "Qwen/Qwen3-14B-FP8": {"repo_id": "Qwen/Qwen3-14B-FP8", "description": "Qwen/Qwen3-14B-FP8"},
     #"Qwen/Qwen3-32B-FP8": {"repo_id": "Qwen/Qwen3-32B-FP8", "description": "Qwen/Qwen3-32B-FP8"},
     "DeepSeek-R1-0528-Qwen3-8B": {"repo_id": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B", "description": "DeepSeek-R1-0528-Qwen3-8B"},