Langfuse SDK

適用於 Langfuse 的轉接端點 - 使用 LiteLLM Virtual Key 呼叫 langfuse 端點。

只要將 https://us.cloud.langfuse.com 替換為 LITELLM_PROXY_BASE_URL/langfuse 即可 🚀

範例用法

from langfuse import Langfuse

langfuse = Langfuse(
    host="http://localhost:4000/langfuse", # your litellm proxy endpoint
    public_key="anything",        # no key required since this is a pass through
    secret_key="LITELLM_VIRTUAL_KEY",        # no key required since this is a pass through
)

print("sending langfuse trace request")
span = langfuse.start_observation(name="test-trace-litellm-proxy-passthrough")
span.end()  # the root observation's name/input/output become the trace's
print("flushing langfuse request")
langfuse.flush()

print("flushed langfuse request")

支援 所有 Langfuse 端點。

查看所有 Langfuse 端點

快速開始

我們來將一個 trace 記錄到 Langfuse。

1. 將 Langfuse 公開/私密金鑰加入環境

export LANGFUSE_PUBLIC_KEY=""
export LANGFUSE_SECRET_KEY=""

2. 啟動 LiteLLM Proxy

litellm

# RUNNING on http://0.0.0.0:4000

3. 測試它！

我們來將一個 trace 記錄到 Langfuse！

from langfuse import Langfuse

langfuse = Langfuse(
    host="http://localhost:4000/langfuse", # your litellm proxy endpoint
    public_key="anything",        # no key required since this is a pass through
    secret_key="anything",        # no key required since this is a pass through
)

print("sending langfuse trace request")
span = langfuse.start_observation(name="test-trace-litellm-proxy-passthrough")
span.end()  # the root observation's name/input/output become the trace's
print("flushing langfuse request")
langfuse.flush()

print("flushed langfuse request")

進階 - 搭配 Virtual Key 使用

前置需求

• 使用 DB 設定 proxy

使用這個方法，可避免將原始 Google AI Studio 金鑰提供給開發人員，同時仍讓他們能使用 Google AI Studio 端點。

用法

1. 設定環境

export DATABASE_URL=""
export LITELLM_MASTER_KEY=""
export LANGFUSE_PUBLIC_KEY=""
export LANGFUSE_SECRET_KEY=""

litellm

# RUNNING on http://0.0.0.0:4000

2. 產生 virtual key

curl -X POST 'http://0.0.0.0:4000/key/generate' \
-H "Authorization: Bearer $LITELLM_API_KEY" \
-H 'Content-Type: application/json' \
-d '{}'

預期回應

{
    ...
    "key": "sk-<virtual-key>"
}

3. 測試它！

from langfuse import Langfuse

langfuse = Langfuse(
    host="http://localhost:4000/langfuse", # your litellm proxy endpoint
    public_key="anything",        # no key required since this is a pass through
    secret_key="sk-<your-litellm-api-key>",        # no key required since this is a pass through
)

print("sending langfuse trace request")
span = langfuse.start_observation(name="test-trace-litellm-proxy-passthrough")
span.end()  # the root observation's name/input/output become the trace's
print("flushing langfuse request")
langfuse.flush()

print("flushed langfuse request")

進階 - 將記錄寫入不同的 langfuse 專案（依金鑰/團隊）