Replicate

LiteLLM 支持 Replicate 上的所有模型

用法

SDK
代理

API 密钥

import os 
os.environ["REPLICATE_API_KEY"] = ""

示例调用

from litellm import completion
import os
## set ENV variables
os.environ["REPLICATE_API_KEY"] = "replicate key"

# replicate llama-3 call
response = completion(
    model="replicate/meta/meta-llama-3-8b-instruct", 
    messages = [{ "content": "Hello, how are you?","role": "user"}]
)

将模型添加到 config.yaml

model_list:
  - model_name: llama-3
    litellm_params:
      model: replicate/meta/meta-llama-3-8b-instruct
      api_key: os.environ/REPLICATE_API_KEY

启动代理

$ litellm --config /path/to/config.yaml --debug

向 LiteLLM 代理服务器发送请求

OpenAI Python v1.0.0+
curl

import openai
client = openai.OpenAI(
    api_key="sk-1234",             # pass litellm proxy key, if you're using virtual keys
    base_url="http://0.0.0.0:4000" # litellm-proxy-base url
)

response = client.chat.completions.create(
    model="llama-3",
    messages = [
      {
          "role": "system",
          "content": "Be a good human!"
      },
      {
          "role": "user",
          "content": "What do you know about earth?"
      }
  ]
)

print(response)

curl --location 'http://0.0.0.0:4000/chat/completions' \
    --header 'Authorization: Bearer sk-1234' \
    --header 'Content-Type: application/json' \
    --data '{
    "model": "llama-3",
    "messages": [
      {
          "role": "system",
          "content": "Be a good human!"
      },
      {
          "role": "user",
          "content": "What do you know about earth?"
      }
      ],
}'

预期的 Replicate 调用

这是 litellm 根据上述示例将向 replicate 发出的调用

POST Request Sent from LiteLLM:
curl -X POST \
https://api.replicate.com/v1/models/meta/meta-llama-3-8b-instruct \
-H 'Authorization: Token your-api-key' -H 'Content-Type: application/json' \
-d '{'version': 'meta/meta-llama-3-8b-instruct', 'input': {'prompt': '<|start_header_id|>system<|end_header_id|>\n\nBe a good human!<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nWhat do you know about earth?<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n'}}'

高级用法 - 提示词格式化

LiteLLM 为所有 meta-llama llama3 instruct 模型提供了提示词模板映射。查看代码

应用自定义提示词模板

SDK
代理

import litellm

import os 
os.environ["REPLICATE_API_KEY"] = ""

# Create your own custom prompt template 
litellm.register_prompt_template(
        model="togethercomputer/LLaMA-2-7B-32K",
        initial_prompt_value="You are a good assistant" # [OPTIONAL]
        roles={
            "system": {
                "pre_message": "[INST] <<SYS>>\n", # [OPTIONAL]
                "post_message": "\n<</SYS>>\n [/INST]\n" # [OPTIONAL]
            },
            "user": { 
                "pre_message": "[INST] ", # [OPTIONAL]
                "post_message": " [/INST]" # [OPTIONAL]
            }, 
            "assistant": {
                "pre_message": "\n" # [OPTIONAL]
                "post_message": "\n" # [OPTIONAL]
            }
        }
        final_prompt_value="Now answer as best you can:" # [OPTIONAL]
)

def test_replicate_custom_model():
    model = "replicate/togethercomputer/LLaMA-2-7B-32K"
    response = completion(model=model, messages=messages)
    print(response['choices'][0]['message']['content'])
    return response

test_replicate_custom_model()

# Model-specific parameters
model_list:
  - model_name: mistral-7b # model alias
    litellm_params: # actual params for litellm.completion()
      model: "replicate/mistralai/Mistral-7B-Instruct-v0.1" 
      api_key: os.environ/REPLICATE_API_KEY
      initial_prompt_value: "\n"
      roles: {"system":{"pre_message":"<|im_start|>system\n", "post_message":"<|im_end|>"}, "assistant":{"pre_message":"<|im_start|>assistant\n","post_message":"<|im_end|>"}, "user":{"pre_message":"<|im_start|>user\n","post_message":"<|im_end|>"}}
      final_prompt_value: "\n"
      bos_token: "<s>"
      eos_token: "</s>"
      max_tokens: 4096

高级用法 - 调用 Replicate 部署

调用已部署的 Replicate LLM 在模型名称前添加 replicate/deployments/ 前缀，这样 litellm 就会调用 deployments 端点。这将调用 replicate 上的 ishaan-jaff/ishaan-mistral 部署

response = completion(
    model="replicate/deployments/ishaan-jaff/ishaan-mistral", 
    messages= [{ "content": "Hello, how are you?","role": "user"}]
)

Replicate 冷启动

由于 Replicate 冷启动，响应可能需要 3-5 分钟。如果你正在调试，请尝试使用 litellm.set_verbose=True 发送请求。更多关于 Replicate 冷启动的信息

Replicate 模型

liteLLM 支持所有 Replicate LLM

对于 Replicate 模型，请确保在 model 参数前添加 replicate/ 前缀。liteLLM 使用此参数检测它。

以下是使用 liteLLM 调用 Replicate LLM 的示例

模型名称	函数调用	必需的操作系统变量
replicate/llama-2-70b-chat	`completion(model='replicate/llama-2-70b-chat:2796ee9483c3fd7aa2e171d38f4ca12251a30609463dcfd4cd76703f22e96cdf', messages)`	`os.environ['REPLICATE_API_KEY']`
a16z-infra/llama-2-13b-chat	`completion(model='replicate/a16z-infra/llama-2-13b-chat:2a7f981751ec7fdf87b5b91ad4db53683a98082e9ff7bfd12c8cd5ea85980a52', messages)`	`os.environ['REPLICATE_API_KEY']`
replicate/vicuna-13b	`completion(model='replicate/vicuna-13b:6282abe6a492de4145d7bb601023762212f9ddbbe78278bd6771c8b3b2f2a13b', messages)`	`os.environ['REPLICATE_API_KEY']`
daanelson/flan-t5-large	`completion(model='replicate/daanelson/flan-t5-large:ce962b3f6792a57074a601d3979db5839697add2e4e02696b3ced4c022d4767f', messages)`	`os.environ['REPLICATE_API_KEY']`
自定义 LLM	`completion(model='replicate/custom-llm-version-id', messages)`	`os.environ['REPLICATE_API_KEY']`
Replicate 部署	`completion(model='replicate/deployments/ishaan-jaff/ishaan-mistral', messages)`	`os.environ['REPLICATE_API_KEY']`

传递附加参数 - max_tokens, temperature

查看所有 litellm.completion 支持的参数此处

# !pip install litellm
from litellm import completion
import os
## set ENV variables
os.environ["REPLICATE_API_KEY"] = "replicate key"

# replicate llama-2 call
response = completion(
    model="replicate/llama-2-70b-chat:2796ee9483c3fd7aa2e171d38f4ca12251a30609463dcfd4cd76703f22e96cdf", 
    messages = [{ "content": "Hello, how are you?","role": "user"}],
    max_tokens=20,
    temperature=0.5
)

代理

  model_list:
    - model_name: llama-3
      litellm_params:
        model: replicate/meta/meta-llama-3-8b-instruct
        api_key: os.environ/REPLICATE_API_KEY
        max_tokens: 20
        temperature: 0.5

传递 Replicate 特定参数

发送 litellm.completion() 不支持，但 Replicate 支持的参数，将其传递给 litellm.completion

示例：seed, min_tokens 是 Replicate 特定参数

# !pip install litellm
from litellm import completion
import os
## set ENV variables
os.environ["REPLICATE_API_KEY"] = "replicate key"

# replicate llama-2 call
response = completion(
    model="replicate/llama-2-70b-chat:2796ee9483c3fd7aa2e171d38f4ca12251a30609463dcfd4cd76703f22e96cdf", 
    messages = [{ "content": "Hello, how are you?","role": "user"}],
    seed=-1,
    min_tokens=2,
    top_k=20,
)

代理

  model_list:
    - model_name: llama-3
      litellm_params:
        model: replicate/meta/meta-llama-3-8b-instruct
        api_key: os.environ/REPLICATE_API_KEY
        min_tokens: 2
        top_k: 20

Replicate

用法​

API 密钥​

示例调用​

预期的 Replicate 调用​

高级用法 - 提示词格式化​

高级用法 - 调用 Replicate 部署​

Replicate 模型​

传递附加参数 - max_tokens, temperature​

传递 Replicate 特定参数​

用法