mirror of
https://github.com/csunny/DB-GPT.git
synced 2025-07-26 05:23:37 +00:00
feat(model): Support Qwen1.5-32B (#1385)
This commit is contained in:
parent
d4da50330f
commit
df36b947d1
@ -158,6 +158,7 @@ At present, we have introduced several key features to showcase our current capa
|
|||||||
We offer extensive model support, including dozens of large language models (LLMs) from both open-source and API agents, such as LLaMA/LLaMA2, Baichuan, ChatGLM, Wenxin, Tongyi, Zhipu, and many more.
|
We offer extensive model support, including dozens of large language models (LLMs) from both open-source and API agents, such as LLaMA/LLaMA2, Baichuan, ChatGLM, Wenxin, Tongyi, Zhipu, and many more.
|
||||||
|
|
||||||
- News
|
- News
|
||||||
|
- 🔥🔥🔥 [Qwen1.5-32B-Chat](https://huggingface.co/Qwen/Qwen1.5-32B-Chat)
|
||||||
- 🔥🔥🔥 [Starling-LM-7B-beta](https://huggingface.co/Nexusflow/Starling-LM-7B-beta)
|
- 🔥🔥🔥 [Starling-LM-7B-beta](https://huggingface.co/Nexusflow/Starling-LM-7B-beta)
|
||||||
- 🔥🔥🔥 [gemma-7b-it](https://huggingface.co/google/gemma-7b-it)
|
- 🔥🔥🔥 [gemma-7b-it](https://huggingface.co/google/gemma-7b-it)
|
||||||
- 🔥🔥🔥 [gemma-2b-it](https://huggingface.co/google/gemma-2b-it)
|
- 🔥🔥🔥 [gemma-2b-it](https://huggingface.co/google/gemma-2b-it)
|
||||||
|
@ -152,6 +152,7 @@
|
|||||||
海量模型支持,包括开源、API代理等几十种大语言模型。如LLaMA/LLaMA2、Baichuan、ChatGLM、文心、通义、智谱等。当前已支持如下模型:
|
海量模型支持,包括开源、API代理等几十种大语言模型。如LLaMA/LLaMA2、Baichuan、ChatGLM、文心、通义、智谱等。当前已支持如下模型:
|
||||||
|
|
||||||
- 新增支持模型
|
- 新增支持模型
|
||||||
|
- 🔥🔥🔥 [Qwen1.5-32B-Chat](https://huggingface.co/Qwen/Qwen1.5-32B-Chat)
|
||||||
- 🔥🔥🔥 [Starling-LM-7B-beta](https://huggingface.co/Nexusflow/Starling-LM-7B-beta)
|
- 🔥🔥🔥 [Starling-LM-7B-beta](https://huggingface.co/Nexusflow/Starling-LM-7B-beta)
|
||||||
- 🔥🔥🔥 [gemma-7b-it](https://huggingface.co/google/gemma-7b-it)
|
- 🔥🔥🔥 [gemma-7b-it](https://huggingface.co/google/gemma-7b-it)
|
||||||
- 🔥🔥🔥 [gemma-2b-it](https://huggingface.co/google/gemma-2b-it)
|
- 🔥🔥🔥 [gemma-2b-it](https://huggingface.co/google/gemma-2b-it)
|
||||||
|
@ -99,6 +99,13 @@ LLM_MODEL_CONFIG = {
|
|||||||
"qwen-1.8b-chat-int8": os.path.join(MODEL_PATH, "wen-1_8B-Chat-Int8"),
|
"qwen-1.8b-chat-int8": os.path.join(MODEL_PATH, "wen-1_8B-Chat-Int8"),
|
||||||
# https://huggingface.co/Qwen/Qwen-1_8B-Chat-Int4
|
# https://huggingface.co/Qwen/Qwen-1_8B-Chat-Int4
|
||||||
"qwen-1.8b-chat-int4": os.path.join(MODEL_PATH, "Qwen-1_8B-Chat-Int4"),
|
"qwen-1.8b-chat-int4": os.path.join(MODEL_PATH, "Qwen-1_8B-Chat-Int4"),
|
||||||
|
# https://huggingface.co/Qwen/Qwen1.5-1.8B-Chat
|
||||||
|
"qwen1.5-1.8b-chat": os.path.join(MODEL_PATH, "Qwen1.5-1.8B-Chat"),
|
||||||
|
"qwen1.5-7b-chat": os.path.join(MODEL_PATH, "Qwen1.5-7B-Chat"),
|
||||||
|
"qwen1.5-14b-chat": os.path.join(MODEL_PATH, "Qwen1.5-14B-Chat"),
|
||||||
|
# https://huggingface.co/Qwen/Qwen1.5-32B-Chat
|
||||||
|
"qwen1.5-32b-chat": os.path.join(MODEL_PATH, "Qwen1.5-32B-Chat"),
|
||||||
|
"qwen1.5-72b-chat": os.path.join(MODEL_PATH, "Qwen1.5-72B-Chat"),
|
||||||
# (Llama2 based) We only support WizardLM-13B-V1.2 for now, which is trained from Llama-2 13b, see https://huggingface.co/WizardLM/WizardLM-13B-V1.2
|
# (Llama2 based) We only support WizardLM-13B-V1.2 for now, which is trained from Llama-2 13b, see https://huggingface.co/WizardLM/WizardLM-13B-V1.2
|
||||||
"wizardlm-13b": os.path.join(MODEL_PATH, "WizardLM-13B-V1.2"),
|
"wizardlm-13b": os.path.join(MODEL_PATH, "WizardLM-13B-V1.2"),
|
||||||
# wget https://huggingface.co/TheBloke/vicuna-13B-v1.5-GGUF/resolve/main/vicuna-13b-v1.5.Q4_K_M.gguf -O models/ggml-model-q4_0.gguf
|
# wget https://huggingface.co/TheBloke/vicuna-13B-v1.5-GGUF/resolve/main/vicuna-13b-v1.5.Q4_K_M.gguf -O models/ggml-model-q4_0.gguf
|
||||||
|
@ -246,8 +246,33 @@ class StarlingLMAdapter(NewHFChatModelAdapter):
|
|||||||
return str_prompt
|
return str_prompt
|
||||||
|
|
||||||
|
|
||||||
|
class QwenAdapter(NewHFChatModelAdapter):
|
||||||
|
"""
|
||||||
|
https://huggingface.co/Qwen/Qwen1.5-32B-Chat
|
||||||
|
|
||||||
|
TODO: There are problems with quantization.
|
||||||
|
"""
|
||||||
|
|
||||||
|
support_4bit: bool = True
|
||||||
|
support_8bit: bool = False # TODO: Support 8bit quantization
|
||||||
|
|
||||||
|
def check_transformer_version(self, current_version: str) -> None:
|
||||||
|
if not current_version >= "4.37.0":
|
||||||
|
raise ValueError(
|
||||||
|
"Qwen 1.5 require transformers.__version__>=4.37.0, please upgrade your transformers package."
|
||||||
|
)
|
||||||
|
|
||||||
|
def do_match(self, lower_model_name_or_path: Optional[str] = None):
|
||||||
|
return (
|
||||||
|
lower_model_name_or_path
|
||||||
|
and "qwen" in lower_model_name_or_path
|
||||||
|
and "1.5" in lower_model_name_or_path
|
||||||
|
)
|
||||||
|
|
||||||
|
|
||||||
register_model_adapter(YiAdapter)
|
register_model_adapter(YiAdapter)
|
||||||
register_model_adapter(Mixtral8x7BAdapter)
|
register_model_adapter(Mixtral8x7BAdapter)
|
||||||
register_model_adapter(SOLARAdapter)
|
register_model_adapter(SOLARAdapter)
|
||||||
register_model_adapter(GemmaAdapter)
|
register_model_adapter(GemmaAdapter)
|
||||||
register_model_adapter(StarlingLMAdapter)
|
register_model_adapter(StarlingLMAdapter)
|
||||||
|
register_model_adapter(QwenAdapter)
|
||||||
|
@ -6,6 +6,7 @@ def is_datetime(value):
|
|||||||
|
|
||||||
|
|
||||||
def convert_datetime_in_row(row):
|
def convert_datetime_in_row(row):
|
||||||
return [value.strftime('%Y-%m-%d %H:%M:%S') if is_datetime(value)
|
return [
|
||||||
else value for value in row]
|
value.strftime("%Y-%m-%d %H:%M:%S") if is_datetime(value) else value
|
||||||
|
for value in row
|
||||||
|
]
|
||||||
|
@ -14,3 +14,4 @@ aioresponses
|
|||||||
pre-commit
|
pre-commit
|
||||||
# Type checking
|
# Type checking
|
||||||
mypy==1.7.0
|
mypy==1.7.0
|
||||||
|
httpx==0.26.0
|
Loading…
Reference in New Issue
Block a user