Merge pull request #2410 from janhq/chore/update-modelhub

Chore: Update model hub v0.4.10

models/command-r-34b/model.json

@@ -0,0 +1,35 @@
+{
+ "sources": [
+ {
+ "filename": "c4ai-command-r-v01-Q4_K_M.gguf",
+ "url": "https://huggingface.co/andrewcanis/c4ai-command-r-v01-GGUF/resolve/main/c4ai-command-r-v01-Q4_K_M.gguf"
+ }
+ ],
+ "id": "command-r-34b",
+ "object": "model",
+ "name": "Command-R v01 34B Q4",
+ "version": "1.0",
+ "description": "C4AI Command-R developed by CohereAI is optimized for a variety of use cases including reasoning, summarization, and question answering.",
+ "format": "gguf",
+ "settings": {
+ "ctx_len": 4096,
+ "prompt_template": "<|START_OF_TURN_TOKEN|><|USER_TOKEN|>{prompt}<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>",
+ "llama_model_path": "c4ai-command-r-v01-Q4_K_M.gguf"
+ },
+ "parameters": {
+ "temperature": 0.7,
+ "top_p": 0.95,
+ "stream": true,
+ "max_tokens": 4096,
+ "stop": [],
+ "frequency_penalty": 0,
+ "presence_penalty": 0
+ },
+ "metadata": {
+ "author": "CohereAI",
+ "tags": ["34B", "Finetuned"],
+ "size": 21500000000
+ },
+ "engine": "nitro"
+ }
+

models/dolphin-phi-2/model.json

@@ -9,7 +9,7 @@
  "object": "model",
  "name": "Dolphin Phi-2 2.7B Q8",
  "version": "1.0",
- "description": "Dolphin Phi-2 is a 2.7B model, fine-tuned for chat, excelling in common sense and logical reasoning benchmarks.",
+ "description": "Dolphin Phi-2 is a good alternative for Phi-2 in chatting",
  "format": "gguf",
  "settings": {
  "ctx_len": 4096,

models/gemma-2b/model.json

@@ -27,7 +27,7 @@
  },
  "metadata": {
  "author": "Google",
- "tags": ["2B", "Finetuned"],
+ "tags": ["2B", "Finetuned", "Tiny"],
  "size": 1500000000
  },
  "engine": "nitro"

models/gemma-7b/model.json

@@ -9,7 +9,7 @@
  "object": "model",
  "name": "Gemma 7B Q4",
  "version": "1.0",
- "description": "Gemma is built from the same technology with Google's Gemini.",
+ "description": "Google's Gemma is built for multilingual purpose",
  "format": "gguf",
  "settings": {
  "ctx_len": 4096,
@@ -27,7 +27,7 @@
  },
  "metadata": {
  "author": "Google",
- "tags": ["7B", "Finetuned"],
+ "tags": ["7B", "Finetuned", "Featured"],
  "size": 5330000000
  },
  "engine": "nitro"

models/hermes-pro-7b/model.json

@@ -0,0 +1,35 @@
+{
+ "sources": [
+ {
+ "filename": "Hermes-2-Pro-Mistral-7B.Q4_K_M.gguf",
+ "url": "https://huggingface.co/NousResearch/Hermes-2-Pro-Mistral-7B-GGUF/resolve/main/Hermes-2-Pro-Mistral-7B.Q4_K_M.gguf"
+ }
+ ],
+ "id": "hermes-pro-7b",
+ "object": "model",
+ "name": "Hermes Pro 7B Q4",
+ "version": "1.0",
+ "description": "Hermes Pro is superior in Roleplaying, Reasoning and Explaining problem.",
+ "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",
+ "llama_model_path": "Hermes-2-Pro-Mistral-7B.Q4_K_M.gguf"
+ },
+ "parameters": {
+ "temperature": 0.7,
+ "top_p": 0.95,
+ "stream": true,
+ "max_tokens": 4096,
+ "stop": [],
+ "frequency_penalty": 0,
+ "presence_penalty": 0
+ },
+ "metadata": {
+ "author": "NousResearch",
+ "tags": ["7B", "Finetuned", "Featured"],
+ "size": 4370000000
+ },
+ "engine": "nitro"
+ }
+

models/llama2-chat-70b-q4/model.json → models/llama2-chat-70b/model.json

@@ -5,11 +5,11 @@
  "url": "https://huggingface.co/TheBloke/Llama-2-70B-Chat-GGUF/resolve/main/llama-2-70b-chat.Q4_K_M.gguf"
  }
  ],
- "id": "llama2-chat-70b-q4",
+ "id": "llama2-chat-70b",
  "object": "model",
  "name": "Llama 2 Chat 70B Q4",
  "version": "1.0",
- "description": "Llama 2 Chat 7b model, specifically designed for a comprehensive understanding through training on extensive internet data.",
+ "description": "Llama 2 specifically designed for a comprehensive understanding the world.",
  "format": "gguf",
  "settings": {
  "ctx_len": 4096,
@@ -26,7 +26,7 @@
  "presence_penalty": 0
  },
  "metadata": {
- "author": "MetaAI, The Bloke",
+ "author": "MetaAI",
  "tags": ["70B", "Foundational Model"],
  "size": 43920000000
  },

models/llama2-chat-7b-q4/model.json → models/llama2-chat-7b/model.json

@@ -5,11 +5,11 @@
  "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",
+ "id": "llama2-chat-7b",
  "object": "model",
  "name": "Llama 2 Chat 7B Q4",
  "version": "1.0",
- "description": "Llama 2 Chat 7b model, specifically designed for a comprehensive understanding through training on extensive internet data.",
+ "description": "Llama 2 specifically designed for a comprehensive understanding the world.",
  "format": "gguf",
  "settings": {
  "ctx_len": 4096,
@@ -26,7 +26,7 @@
  "presence_penalty": 0
  },
  "metadata": {
- "author": "MetaAI, The Bloke",
+ "author": "MetaAI",
  "tags": ["7B", "Foundational Model"],
  "size": 4080000000
  },

models/llava-13b/model.json

@@ -0,0 +1,35 @@
+{
+ "sources": [
+ {
+ "filename": "llava-v1.6-vicuna-13b.Q4_K_M.gguf",
+ "url": "https://huggingface.co/cjpais/llava-v1.6-vicuna-13b-gguf/resolve/main/llava-v1.6-vicuna-13b.Q4_K_M.gguf"
+ },
+ {
+ "filename": "mmproj-model-f16.gguf",
+ "url": "https://huggingface.co/cjpais/llava-v1.6-vicuna-13b-gguf/resolve/main/mmproj-model-f16.gguf"
+ }
+ ],
+ "id": "llava-13b",
+ "object": "model",
+ "name": "LlaVa 13B Q4",
+ "version": "1.1",
+ "description": "LlaVa can bring vision understanding to Jan",
+ "format": "gguf",
+ "settings": {
+ "vision_model": true,
+ "text_model": false,
+ "ctx_len": 4096,
+ "prompt_template": "\n### Instruction:\n{prompt}\n### Response:\n",
+ "llama_model_path": "llava-v1.6-vicuna-13b.Q4_K_M.gguf",
+ "mmproj": "mmproj-model-f16.gguf"
+ },
+ "parameters": {
+ "max_tokens": 4096
+ },
+ "metadata": {
+ "author": "liuhaotian",
+ "tags": ["Vision"],
+ "size": 7870000000
+ },
+ "engine": "nitro"
+}

models/llava-7b/model.json

@@ -0,0 +1,35 @@
+{
+ "sources": [
+ {
+ "filename": "llava-v1.6-mistral-7b.Q4_K_M.gguf",
+ "url": "https://huggingface.co/cjpais/llava-1.6-mistral-7b-gguf/resolve/main/llava-v1.6-mistral-7b.Q4_K_M.gguf"
+ },
+ {
+ "filename": "mmproj-model-f16.gguf",
+ "url": "https://huggingface.co/cjpais/llava-1.6-mistral-7b-gguf/resolve/main/mmproj-model-f16.gguf"
+ }
+ ],
+ "id": "llava-7b",
+ "object": "model",
+ "name": "LlaVa 7B",
+ "version": "1.1",
+ "description": "LlaVa can bring vision understanding to Jan",
+ "format": "gguf",
+ "settings": {
+ "vision_model": true,
+ "text_model": false,
+ "ctx_len": 4096,
+ "prompt_template": "\n### Instruction:\n{prompt}\n### Response:\n",
+ "llama_model_path": "llava-v1.6-mistral-7b.Q4_K_M.gguf",
+ "mmproj": "mmproj-model-f16.gguf"
+ },
+ "parameters": {
+ "max_tokens": 4096
+ },
+ "metadata": {
+ "author": "liuhaotian",
+ "tags": ["Vision"],
+ "size": 4370000000
+ },
+ "engine": "nitro"
+}