Chore/update model hub (janhq#1342)

* fix(mistral-ins): clean redundant parameters

* add(yarn-mistral): update new requested model

* fix(trinity-v1): delete trinity v1 from the hub

* add(tulu-2-70b): llama 70b alternative

* fix(lzlv-70b): delete lzlv-70b and changed to tulu-2

* fix(mistral-ins): upgrade model version to v0.2

* fix(model-extention): pump version to 1.0.18

* add(dolphin 8x7b): update the current  best moe finetuned model

* add(openchat): the best 7b model

* fix(tinyllama): pump version of the model to v1

* fix(stealth): upgrade stealth to v1.3

* Revert "fix(stealth): upgrade stealth to v1.3"

This reverts commit da24df3.

* fix(stealth): upgrade version to v1.3

extensions/model-extension/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@janhq/model-extension",
-  "version": "1.0.17",
+  "version": "1.0.18",
   "description": "Model Management Extension provides model exploration and seamless downloads",
   "main": "dist/index.js",
   "module": "dist/module.js",

models/dolphin-2.7-mixtral-8x7b/model.json

@@ -0,0 +1,22 @@
+{
+    "source_url": "https://huggingface.co/TheBloke/dolphin-2.7-mixtral-8x7b-GGUF/resolve/main/dolphin-2.7-mixtral-8x7b.Q4_K_M.gguf",
+    "id": "dolphin-2.7-mixtral-8x7b",
+    "object": "model",
+    "name": "Dolphin 8x7B Q4",
+    "version": "1.0",
+    "description": "This model is an uncensored model based on Mixtral-8x7b. Dolphin is really good at coding",
+    "format": "gguf",
+    "settings": {
+      "ctx_len": 4096,
+      "prompt_template": "<|im_start|>system\n{system_message}<|im_end|>\n<|im_start|>user\n{prompt}<|im_end|>\n<|im_start|>assistant"
+    },
+    "parameters": {
+      "max_tokens": 4096
+    },
+    "metadata": {
+      "author": "Cognitive Computations, TheBloke",
+      "tags": ["70B", "Fintuned"],
+      "size": 26440000000
+    },
+    "engine": "nitro"
+  }

models/mistral-ins-7b-q4/model.json

@@ -1,5 +1,5 @@
 {
-    "source_url": "https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.1-GGUF/resolve/main/mistral-7b-instruct-v0.1.Q4_K_M.gguf",
+    "source_url": "https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GGUF/resolve/main/mistral-7b-instruct-v0.2.Q4_K_M.gguf",
     "id": "mistral-ins-7b-q4",
     "object": "model",
     "name": "Mistral Instruct 7B Q4",
@@ -8,9 +8,6 @@
     "format": "gguf",
     "settings": {
       "ctx_len": 4096,
-      "system_prompt": "",
-      "user_prompt": "<s>[INST]",
-      "ai_prompt": "[/INST]",
       "prompt_template": "<s>[INST]{prompt}\n[/INST]"
     },
     "parameters": {

models/openchat-3.5-7b/model.json

@@ -0,0 +1,22 @@
+{
+    "source_url": "https://huggingface.co/TheBloke/openchat-3.5-1210-GGUF/resolve/main/openchat-3.5-1210.Q4_K_M.gguf",
+    "id": "openchat-3.5-7b",
+    "object": "model",
+    "name": "Openchat-3.5 7B Q4",
+    "version": "1.0",
+    "description": "The performance of this open-source model surpasses that of ChatGPT-3.5 and Grok-1 across various benchmarks.",
+    "format": "gguf",
+    "settings": {
+      "ctx_len": 4096,
+      "prompt_template": "GPT4 Correct User: {prompt}<|end_of_turn|>GPT4 Correct Assistant:"
+    },
+    "parameters": {
+      "max_tokens": 4096
+    },
+    "metadata": {
+      "author": "Openchat",
+      "tags": ["7B", "Fintuned", "Featured"],
+      "size": 4370000000
+    },
+    "engine": "nitro"
+  }

models/stealth-v1.2-7b/model.json

@@ -1,5 +1,5 @@
 {
-    "source_url": "https://huggingface.co/janhq/stealth-v1.2-GGUF/resolve/main/stealth-v1.2.Q4_K_M.gguf",
+    "source_url": "https://huggingface.co/janhq/stealth-v1.3-GGUF/resolve/main/stealth-v1.3.Q4_K_M.gguf",
     "id": "stealth-v1.2-7b",
     "object": "model",
     "name": "Stealth-v1.2 7B Q4",

models/tinyllama-1.1b/model.json

@@ -1,5 +1,5 @@
 {
-  "source_url": "https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat-v0.6/resolve/main/ggml-model-q4_0.gguf",
+  "source_url": "https://huggingface.co/TheBloke/TinyLlama-1.1B-Chat-v1.0-GGUF/resolve/main/tinyllama-1.1b-chat-v1.0.Q4_K_M.gguf",
   "id": "tinyllama-1.1b",
   "object": "model",
   "name": "TinyLlama Chat 1.1B Q4",
@@ -16,7 +16,7 @@
   "metadata": {
       "author": "TinyLlama",
       "tags": ["Tiny", "Foundation Model"],
-      "size": 637000000
+      "size": 669000000
   },
   "engine": "nitro"
 }

models/tulu-2-70b/model.json

@@ -0,0 +1,22 @@
+{
+    "source_url": "https://huggingface.co/TheBloke/tulu-2-dpo-70B-GGUF/resolve/main/tulu-2-dpo-70b.Q4_K_M.gguf",
+    "id": "tulu-2-70b",
+    "object": "model",
+    "name": "Tulu 2 70B Q4",
+    "version": "1.0",
+    "description": "Tulu V2 DPO 70B is a fine-tuned version of Llama 2 using (DPO). This model is a strong alternative to Llama 2 70b Chat to act as helpful assistants.",
+    "format": "gguf",
+    "settings": {
+      "ctx_len": 4096,
+      "prompt_template": "<|user|>\n{prompt}\n<|assistant|>"
+    },
+    "parameters": {
+      "max_tokens": 4096
+    },
+    "metadata": {
+      "author": "Lizpreciatior, The Bloke",
+      "tags": ["70B", "Finetuned"],
+      "size": 41400000000
+    },
+    "engine": "nitro"
+  }

models/yarn-mistral-7b/model.json

@@ -0,0 +1,23 @@
+{
+    "source_url": "https://huggingface.co/TheBloke/Yarn-Mistral-7B-128k-GGUF/resolve/main/yarn-mistral-7b-128k.Q4_K_M.gguf",
+    "id": "yarn-mistral-7b",
+    "object": "model",
+    "name": "Yarn Mistral 7B Q4",
+    "version": "1.0",
+    "description": "Yarn Mistral 7B is a language model for long context and supports a 128k token context window.",
+    "format": "gguf",
+    "settings": {
+      "ctx_len": 4096,
+      "prompt_template": "{prompt}"
+    },
+    "parameters": {
+      "max_tokens": 4096
+    },
+    "metadata": {
+      "author": "NousResearch, The Bloke",
+      "tags": ["7B","Finetuned"],
+      "size": 4370000000
+    },
+    "engine": "nitro"
+  }
+