36 lines
1.1 KiB
JSON
36 lines
1.1 KiB
JSON
{
|
|
"sources": [
|
|
{
|
|
"filename": "Qwen2.5-72B-Instruct-Q4_K_M.gguf",
|
|
"url": "https://huggingface.co/bartowski/Qwen2.5-72B-Instruct-GGUF/resolve/main/Qwen2.5-72B-Instruct-Q4_K_M.gguf"
|
|
}
|
|
],
|
|
"id": "qwen2.5-72b-instruct",
|
|
"object": "model",
|
|
"name": "Qwen2.5 72B Instruct Q4",
|
|
"version": "1.0",
|
|
"description": "The Qwen 2.5 family is specifically designed to excel in math, coding, and other computational tasks",
|
|
"format": "gguf",
|
|
"settings": {
|
|
"ctx_len": 32768,
|
|
"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": "Qwen2.5-72B-Instruct-Q4_K_M.gguf",
|
|
"ngl": 81
|
|
},
|
|
"parameters": {
|
|
"temperature": 0.7,
|
|
"top_p": 0.95,
|
|
"stream": true,
|
|
"max_tokens": 32768,
|
|
"stop": ["<|endoftext|>", "<|im_end|>"],
|
|
"frequency_penalty": 0,
|
|
"presence_penalty": 0
|
|
},
|
|
"metadata": {
|
|
"author": "QwenLM",
|
|
"tags": ["72B"],
|
|
"size": 47400000000
|
|
},
|
|
"engine": "nitro"
|
|
}
|
|
|