-
Notifications
You must be signed in to change notification settings - Fork 2
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
4090 24G显存 单卡是否可以运行? #1
Comments
您好,很高兴你能认可我们的工作。该基准及框架是通过api调用大模型,因此对显卡性能并没有过高的要求,完全可以部署在4090上! |
我调用deepseek的api运行的时间很慢是什么原因?请问运行eval_LLM.py文件正常需要多少时间? def call_with_messages_deepseek(QA, save_list, model="deepseek-chat"): File:hyperedge_count_cot_bag_test:0/5: 0%| | 2/3500 [01:20<39:51:26, 41.02s/it] |
通过我们近期的相关工作来看,deepseek的推理速度很慢...虽然说他的深度思考确实提升了模型的性能,但是时间消耗极大,几乎是别的大模型的十几倍。
如果你想快速的看到结果,那么我建议你使用gpt-4o-mini,国产大模型中qwen系列也还不错,deepseek建议在时间充裕的情况下使用。
gpt-4o-mini的eval很快,进度条几乎是飞一般的加载,其他大模型得看时间段,有的时候很快,有的时候又慢的让人窒息,deepseek在晚上非高峰时间段使用速度也勉强能接受。
超越
***@***.***
…------------------ 原始邮件 ------------------
发件人: "iMoonLab/LLM4Hypergraph" ***@***.***>;
发送时间: 2025年3月10日(星期一) 上午10:40
***@***.***>;
***@***.******@***.***>;
主题: Re: [iMoonLab/LLM4Hypergraph] 4090 24G显存 单卡是否可以运行? (Issue #1)
我调用deepseek的api运行的时间很慢是什么原因?请问运行eval_LLM.py文件正常需要多少时间?
def call_with_messages_deepseek(QA, save_list, model="deepseek-chat"):
api_key = ""
client = OpenAI(api_key=api_key, base_url="https://api.deepseek.com/v1")
question = QA['question']
# print(question)
while True:
try:
response = client.chat.completions.create(
model=model,
messages=[{"role": "user", "content": question}],
temperature=0.7,
max_tokens=1024,
stream=False
)
QA.update({'response': response.choices[0].message.content, 'result': True})
save_list.append(QA)
break
except Exception as e:
print(f"DeepSeek 请求失败: {e}")
time.sleep(2)
File:hyperedge_count_cot_bag_test:0/5: 0%| | 2/3500 [01:20<39:51:26, 41.02s/it]
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you commented.Message ID: ***@***.***>
yuwxyun275 left a comment (iMoonLab/LLM4Hypergraph#1)
我调用deepseek的api运行的时间很慢是什么原因?请问运行eval_LLM.py文件正常需要多少时间?
def call_with_messages_deepseek(QA, save_list, model="deepseek-chat"):
api_key = ""
client = OpenAI(api_key=api_key, base_url="https://api.deepseek.com/v1")
question = QA['question']
# print(question)
while True:
try:
response = client.chat.completions.create(
model=model,
messages=[{"role": "user", "content": question}],
temperature=0.7,
max_tokens=1024,
stream=False
)
QA.update({'response': response.choices[0].message.content, 'result': True})
save_list.append(QA)
break
except Exception as e:
print(f"DeepSeek 请求失败: {e}")
time.sleep(2)
File:hyperedge_count_cot_bag_test:0/5: 0%| | 2/3500 [01:20<39:51:26, 41.02s/it]
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you commented.Message ID: ***@***.***>
|
作者您好,我对您的研究工作很感兴趣并且很感谢您将代码开源。我想询问单张24G显存的4090是否可以部署该模型?
The text was updated successfully, but these errors were encountered: