Update constants.py

constants.py

@@ -61,13 +61,13 @@ DOCUMENT_MAP = {
 }
 
 # Default Instructor Model
-EMBEDDING_MODEL_NAME = "hkunlp/instructor-large"  # Uses 1.5 GB of VRAM (High Accuracy with lower VRAM usage)
+# EMBEDDING_MODEL_NAME = "hkunlp/instructor-large"  # Uses 1.5 GB of VRAM (High Accuracy with lower VRAM usage)
 
 ####
 #### OTHER EMBEDDING MODEL OPTIONS
 ####
 
-# EMBEDDING_MODEL_NAME = "hkunlp/instructor-xl" # Uses 5 GB of VRAM (Most Accurate of all models)
+EMBEDDING_MODEL_NAME = "hkunlp/instructor-xl" # Uses 5 GB of VRAM (Most Accurate of all models)
 # EMBEDDING_MODEL_NAME = "intfloat/e5-large-v2" # Uses 1.5 GB of VRAM (A little less accurate than instructor-large)
 # EMBEDDING_MODEL_NAME = "intfloat/e5-base-v2" # Uses 0.5 GB of VRAM (A good model for lower VRAM GPUs)
 # EMBEDDING_MODEL_NAME = "all-MiniLM-L6-v2" # Uses 0.2 GB of VRAM (Less accurate but fastest - only requires 150mb of vram)
@@ -100,11 +100,11 @@ EMBEDDING_MODEL_NAME = "hkunlp/instructor-large"  # Uses 1.5 GB of VRAM (High Ac
 #### (FOR GGUF MODELS)
 ####
 
-MODEL_ID = "TheBloke/Llama-2-13b-Chat-GGUF"
-MODEL_BASENAME = "llama-2-13b-chat.Q4_K_M.gguf"
+# MODEL_ID = "TheBloke/Llama-2-13b-Chat-GGUF"
+# MODEL_BASENAME = "llama-2-13b-chat.Q4_K_M.gguf"
 
-# MODEL_ID = "TheBloke/Llama-2-7b-Chat-GGUF"
-# MODEL_BASENAME = "llama-2-7b-chat.Q4_K_M.gguf"
+MODEL_ID = "TheBloke/Llama-2-7b-Chat-GGUF"
+MODEL_BASENAME = "llama-2-7b-chat.Q4_K_M.gguf"
 
 # MODEL_ID = "TheBloke/Mistral-7B-Instruct-v0.1-GGUF"
 # MODEL_BASENAME = "mistral-7b-instruct-v0.1.Q8_0.gguf"