diff --git a/models/capybara-34b/model.json b/models/capybara-34b/model.json
index d2da8d002..7c669bbb7 100644
--- a/models/capybara-34b/model.json
+++ b/models/capybara-34b/model.json
@@ -7,13 +7,13 @@
     "description": "Nous Capybara 34B, a variant of the Yi-34B model, is the first Nous model with a 200K context length, trained for three epochs on the innovative Capybara dataset.",
     "format": "gguf",
     "settings": {
-      "ctx_len": 4096,
+      "ctx_len": 2048,
       "system_prompt": "",
       "user_prompt": "USER: ",
       "ai_prompt": "ASSISTANT: "
     },
     "parameters": {
-      "max_tokens": 4096
+      "max_tokens": 2048
     },
     "metadata": {
       "author": "NousResearch, The Bloke",
diff --git a/models/deepseek-coder-1.3b/model.json b/models/deepseek-coder-1.3b/model.json
index 3ee705d20..dac90423d 100644
--- a/models/deepseek-coder-1.3b/model.json
+++ b/models/deepseek-coder-1.3b/model.json
@@ -7,13 +7,13 @@
     "description": "",
     "format": "gguf",
     "settings": {
-      "ctx_len": 4096,
+      "ctx_len": 2048,
       "system_prompt": "",
       "user_prompt": "",
       "ai_prompt": ""
     },
     "parameters": {
-      "max_tokens": 4096
+      "max_tokens": 2048
     },
     "metadata": {
       "author": "deepseek, The Bloke",
diff --git a/models/llama2-chat-7b-q4/model.json b/models/llama2-chat-7b-q4/model.json
new file mode 100644
index 000000000..be302d144
--- /dev/null
+++ b/models/llama2-chat-7b-q4/model.json
@@ -0,0 +1,24 @@
+{
+    "source_url": "https://huggingface.co/TheBloke/Llama-2-7B-Chat-GGUF/resolve/main/llama-2-7b-chat.Q4_K_M.gguf",
+    "id": "llama2-chat-7b-q4",
+    "object": "model",
+    "name": "Llama 2 Chat 7B Q4",
+    "version": "1.0",
+    "description": "This is a 4-bit quantized version of Meta AI's Llama 2 Chat 7b model.",
+    "format": "gguf",
+    "settings": {
+      "ctx_len": 2048,
+      "system_prompt": "[INST] <<SYS>>\n",
+      "user_prompt": "<</SYS>>\n",
+      "ai_prompt": "[/INST]"
+    },
+    "parameters": {
+      "max_tokens": 2048
+    },
+    "metadata": {
+      "author": "MetaAI, The Bloke",
+      "tags": ["Foundational Model", "General", "Code"],
+      "size": 4080000000
+    }
+  }
+  
\ No newline at end of file
diff --git a/models/llama2-chat-7b-q5/model.json b/models/llama2-chat-7b-q5/model.json
new file mode 100644
index 000000000..8a93327d6
--- /dev/null
+++ b/models/llama2-chat-7b-q5/model.json
@@ -0,0 +1,24 @@
+{
+    "source_url": "https://huggingface.co/TheBloke/Llama-2-7B-Chat-GGUF/resolve/main/llama-2-7b-chat.Q5_K_M.gguf",
+    "id": "llama2-chat-7b-q5",
+    "object": "model",
+    "name": "Llama 2 Chat 7B Q5",
+    "version": "1.0",
+    "description": "This is a 5-bit quantized version of Meta AI's Llama 2 Chat 7b model.",
+    "format": "gguf",
+    "settings": {
+      "ctx_len": 2048,
+      "system_prompt": "[INST] <<SYS>>\n",
+      "user_prompt": "<</SYS>>\n",
+      "ai_prompt": "[/INST]"
+    },
+    "parameters": {
+      "max_tokens": 2048
+    },
+    "metadata": {
+      "author": "MetaAI, The Bloke",
+      "tags": ["Foundational Model", "General", "Code"],
+      "size": 4780000000
+    }
+  }
+  
\ No newline at end of file