Skip to main content
本指南介绍如何加载 Volcano Embedding 类。

​
API 初始化

要使用基于 VolcEngine 的 LLM 服务，你需要初始化以下参数： 你可以选择在环境变量中初始化 AK、SK，也可以在初始化参数中直接传入： 
export VOLC_ACCESSKEY=XXX
export VOLC_SECRETKEY=XXX

"""For basic init and call"""
import os

from langchain_community.embeddings import VolcanoEmbeddings

os.environ["VOLC_ACCESSKEY"] = ""
os.environ["VOLC_SECRETKEY"] = ""

embed = VolcanoEmbeddings(volcano_ak="", volcano_sk="")
print("embed_documents result:")
res1 = embed.embed_documents(["foo", "bar"])
for r in res1:
    print("", r[:8])

embed_documents result:
 [0.02929673343896866, -0.009310632012784481, -0.060323506593704224, 0.0031018739100545645, -0.002218986628577113, -0.0023125179577618837, -0.04864659160375595, -2.062115163425915e-05]
 [0.01987231895327568, -0.026041055098176003, -0.08395249396562576, 0.020043574273586273, -0.028862033039331436, 0.004629664588719606, -0.023107370361685753, -0.0342753604054451]

print("embed_query result:")
res2 = embed.embed_query("foo")
print("", r[:8])

embed_query result:
 [0.01987231895327568, -0.026041055098176003, -0.08395249396562576, 0.020043574273586273, -0.028862033039331436, 0.004629664588719606, -0.023107370361685753, -0.0342753604054451]
Edit this page on GitHub or file an issue.
Connect these docs to Claude, VSCode, and more via MCP for real-time answers.