add q2_k precsion weights

app.py

@@ -18,7 +18,7 @@ MODELS = {
         "clip_repo":  "ggml-org/SmolVLM2-256M-Video-Instruct-GGUF",
         "model_prefix": "SmolVLM2-256M-Video-Instruct",
         "clip_prefix":  "mmproj-SmolVLM2-256M-Video-Instruct",
-        "model_variants": ["Q8_0", "f16"],
+        "model_variants": ["Q2_K","Q8_0", "f16"],
         "clip_variants":  ["Q8_0", "f16"],
     },
     "500M": {
@@ -26,7 +26,7 @@ MODELS = {
         "clip_repo":  "ggml-org/SmolVLM2-500M-Video-Instruct-GGUF",
         "model_prefix": "SmolVLM2-500M-Video-Instruct",
         "clip_prefix":  "mmproj-SmolVLM2-500M-Video-Instruct",
-        "model_variants": ["Q8_0", "f16"],
+        "model_variants": ["Q2_K","Q8_0", "f16"],
         "clip_variants":  ["Q8_0", "f16"],
     },
     "2.2B": {
@@ -34,7 +34,7 @@ MODELS = {
         "clip_repo":  "ggml-org/SmolVLM2-2.2B-Instruct-GGUF",
         "model_prefix": "SmolVLM2-2.2B-Instruct",
         "clip_prefix":  "mmproj-SmolVLM2-2.2B-Instruct",
-        "model_variants": ["Q4_K_M", "Q8_0", "f16"],
+        "model_variants": ["Q2_K","Q4_K_M", "Q8_0", "f16"],
         "clip_variants":  ["Q8_0", "f16"],
     },
 }