# Maxim SDK : Langchain wrapper

<div style="display: flex; justify-content: center; align-items: center;margin-bottom:20px;">
<img src="https://cdn.getmaxim.ai/third-party/langchain-tracer.png">
</div>

This is the langchain wrapper built for [Maxim JS SDK](https://www.npmjs.com/package/@maximai/maxim-js).

## Simplifies integrating Maxim logger with langchain

### Install

```sh
npm install @maximai/maxim-js-langchain
```

### Initialize Maxim logger

```js
const maxim = new Maxim({apiKey: "maxim-api-key"});
const logger = await maxim.logger({ id: "log-repository-id" });
```

### Start logging

```js
const maximTracer = new MaximLangchainTracer(logger);
const llm = new ChatOpenAI({
    openAIApiKey: openAIKey,
    modelName: "gpt-4o",
    temperature: 0,
    callbacks: [maximTracer],
    metadata: {
        maxim: { // optional: add metadata to the generation
            generationName: "maths-gen",
            generationTags: { tag: "test" }
        }
    }
});

const query = "What's the sum of 3 and 2?";

const result = await llm.invoke(query);
```

#### Langchain module compatibility

| | Anthropic | Bedrock Anthropic | Bedrock Meta | OpenAI | Azure |
|---|---|---|---|---|---|
|Chat (0.3.x) | ✅ | ✅ | ✅ | ✅ | ✅ |
|Tool call (0.3.x)| ✅ | ✅ | ⛔️ | ✅ | ✅ |
|Chain (via LLM) (0.3.x)| ✅ | ✅ | ✅ | ✅ | ✅ |
|Streaming (0.3.x)| ✅ | ✅ | ✅ | ✳️ | ✳️ |
|Agent (0.3.x)| ⛔️ | ⛔️ | ⛔️ | ⛔️ | ⛔️ |

✳️ Token usage is not supported by Langchain

## Version changelog

### 1.1.0
- fix: prevention of creation of multiple traces for a single langchain call

### 1.0.0

- Feat: adds LangGraph support

### 0.2.7

- Feat: add support for tool call logging via callbacks

### 0.2.6

- Fix: token usage capturing for anthropic models

### 0.2.5

- Fix: model name and provider parsing while using Langchain

### 0.2.4

- Fix: Peer dependency changed from langchain to @langchain/core
- Chore: tsconfig target moved to es2017

### 0.2.3

- Feat: Updates peer dep for Maxim SDK

### 0.2.2

- Feat: Adds support for langchain 3.x

### 0.2.1

- Improvement: easy logging of retrieval and generation

### 0.1.1

- Breaking: Dropping support for Langchain versions below 0.3.x

### 0.0.4

- Improvement: Changed pining for core Maxim SDK

### 0.0.3

- Fix: Token usage were not getting captured for streaming responses

### 0.0.2

- Fix: Adds empty tags as {}

### 0.0.1

- Early preview