zhuchaowe/maxkb
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: 可选模型去掉无法下载的模型

Browse Source
This commit is contained in:
shaohuzhang1 2024-03-22 18:03:36 +08:00
parent d3aa6463d8
commit 633d81b3fb
1 changed files with 1 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
apps/setting/models_provider/impl/ollama_model_provider/ollama_model_provider.py
View File

@ -76,15 +76,7 @@ model_dict = {
'llama2-chinese:13b': ModelInfo( 'llama2-chinese:13b': ModelInfo(
'llama2-chinese:13b', 'llama2-chinese:13b',
'由于Llama2本身的中文对齐较弱，我们采用中文指令集，对meta-llama/Llama-2-13b-chat-hf进行LoRA微调，使其具备较强的中文对话能力。', '由于Llama2本身的中文对齐较弱，我们采用中文指令集，对meta-llama/Llama-2-13b-chat-hf进行LoRA微调，使其具备较强的中文对话能力。',
ModelTypeConst.LLM, ollama_llm_model_credential), ModelTypeConst.LLM, ollama_llm_model_credential)
'llama2-chinese:13b-maxkb': ModelInfo(
'llama2-chinese:13b-maxkb',
'由于Llama2本身的中文对齐较弱，我们采用中文指令集，对meta-llama/Llama-2-13b-chat-hf进行LoRA微调，使其具备较强的中文对话能力。fi2cloud专用',
ModelTypeConst.LLM, ollama_llm_model_credential),
'baichuan2:13b-chat': ModelInfo(
'baichuan2:13b-chat',
'Baichuan 2 是百川智能推出的新一代开源大语言模型，采用 2.6 万亿 Tokens 的高质量语料训练，在权威的中文和英文 benchmark 上均取得同尺寸最好的效果',
ModelTypeConst.LLM, ollama_llm_model_credential),
} }