Anthropic, Mistral, etc.

With Literal’s integration capabilities, you can seamlessly monitor all the providers supported by Langchain. This allows for a unified observability experience across different AI models and services, ensuring that you can track and analyze the performance and usage of each provider within the Literal platform. You can thus monitor Anthropic, Mistral, and many other LLM providers.

import os, asyncio
from literalai import LiteralClient

from langchain_anthropic import ChatAnthropic
from langchain_core.messages import HumanMessage
from langchain.schema.runnable.config import RunnableConfig

from dotenv import load_dotenv
load_dotenv()


literal_client = LiteralClient(api_key=os.getenv("LITERAL_API_KEY"))
cb = literal_client.langchain_callback()


model = ChatAnthropic(model='claude-3-opus-20240229', anthropic_api_key=os.getenv("ANTHROPIC_API_KEY"))
message = HumanMessage(content="What is the capital of France?")
response = model.invoke([message], config=RunnableConfig(callbacks=[cb]))

literal_client.flush_and_stop()

import { LiteralClient } from "@literalai/client";

import { HumanMessage } from "@langchain/core/messages";
import { ChatAnthropic } from "@langchain/anthropic";

const literal_client = new LiteralClient(process.env.LITERAL_API_KEY);
const cb = literal_client.instrumentation.langchain.literalCallback();

// Example of using the callback
const message = new HumanMessage({ content: "What is the capital of France?" });

const model = new ChatAnthropic({
  anthropicApiKey: process.env.ANTHROPIC_API_KEY,
});
model.invoke([message], {
  callbacks: [cb],
}).then(response => {
  console.log(response)
}).catch(error => {
  // Handle any errors here
});

Get Started

Concepts

Integrations

Examples

Self Hosting

Anthropic, Mistral, etc.