1e0d4f3753
* fix: correct phi3 * redundant phi2 dolphin * add: hermes llama3 * add: ngl settings * correct ctx len * correct ngl * correct maxlen + ngl * disable phi3 * add ngl * add ngl * add ngl * add ngl * add ngl * add ngl * add ngl * remove redundant hermes pro * add ngl * add ngl * add ngl * remove miqu * add ngl * add ngl * add ngl * add ngl * remove redundant * add ngl * add ngl * add ngl * add ngl * add ngl * add ngl * add ngl * add ngl * add ngl * version package bump * feat: resolve issue of cannot found model in the extensions due to the removal * feat: completely remove hermes-pro-7b * feat: completely remove openhermes-neural-7b and miqu-70b, and add llama3-hermes-8b via renaming from Rex * fix: correct description --------- Co-authored-by: Van-QA <van@jan.ai>
37 lines
1.1 KiB
JSON
37 lines
1.1 KiB
JSON
{
|
|
"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": "1.4",
|
|
"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": 131072,
|
|
"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",
|
|
"ngl": 40
|
|
},
|
|
"parameters": {
|
|
"temperature": 0.7,
|
|
"top_p": 0.95,
|
|
"stream": true,
|
|
"max_tokens": 131072,
|
|
"stop": [],
|
|
"frequency_penalty": 0,
|
|
"presence_penalty": 0
|
|
},
|
|
"metadata": {
|
|
"author": "CohereAI",
|
|
"tags": ["34B", "Finetuned", "Featured"],
|
|
"size": 21500000000
|
|
},
|
|
"engine": "nitro"
|
|
}
|
|
|