-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy path.env.example
More file actions
34 lines (29 loc) · 1.25 KB
/
Copy path.env.example
File metadata and controls
34 lines (29 loc) · 1.25 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
# ============================================================
# LangChain 教程 — 环境配置
# ============================================================
# 复制此文件为 .env 并填入你的 API 密钥:
# cp .env.example .env
#
# 本教程默认使用 DeepSeek API(兼容 OpenAI 格式)。
# 如需使用其他提供商,见 README.md 中的"切换 LLM 提供商"。
# 必需:你的 DeepSeek API 密钥
# 在 https://platform.deepseek.com/api_keys 获取
DEEPSEEK_API_KEY=sk-your-deepseek-key-here
# 可选:DeepSeek 模型选择
# deepseek-chat — DeepSeek-V3(默认,性价比最高)
# deepseek-reasoner — DeepSeek-R1(推理增强)
DEEPSEEK_MODEL=deepseek-chat
# ============================================================
# 使用其他提供商(可选)
# ============================================================
# 如果你想用 OpenAI 而非 DeepSeek,取消以下注释:
# OPENAI_API_KEY=sk-your-openai-key-here
# OPENAI_BASE_URL=https://api.openai.com/v1
# LLM_MODEL=gpt-4o-mini
# 使用本地 Ollama(免费):
# OPENAI_BASE_URL=http://localhost:11434/v1
# LLM_MODEL=llama3.2
# 可选:LangSmith 追踪
# LANGCHAIN_TRACING_V2=true
# LANGCHAIN_API_KEY=ls__your-key-here
# LANGCHAIN_PROJECT=langchain-tutorial