Add Granite-4.0 1B model for extraction

- Added granite_4_0_1b_q4 from unsloth/granite-4.0-h-1b-GGUF
- Q4_0 quantization for balanced quality/speed
- 32K context window
- Temperature 0.1 for focused extraction
- Larger model should help with extraction focus compared to 350M

app.py

@@ -801,6 +801,22 @@ EXTRACTION_MODELS = {
             "repeat_penalty": 1.0,
         },
     },
+    "granite_4_0_1b_q4": {
+        "name": "Granite-4.0 1B (32K Context)",
+        "repo_id": "unsloth/granite-4.0-h-1b-GGUF",
+        "filename": "*Q4_0.gguf",
+        "max_context": 32768,
+        "default_n_ctx": 4096,
+        "params_size": "1B",
+        "supports_reasoning": False,
+        "supports_toggle": False,
+        "inference_settings": {
+            "temperature": 0.1,
+            "top_p": 0.95,
+            "top_k": 30,
+            "repeat_penalty": 1.0,
+        },
+    },
     "falcon_h1_1.5b_q4": {
         "name": "Falcon-H1 1.5B Q4",
         "repo_id": "unsloth/Falcon-H1-1.5B-Deep-Instruct-GGUF",