Framework Integrations

ClawPipe integrates with popular AI frameworks. Use it as a drop-in replacement for OpenAI or plug it into LangChain, LlamaIndex, and the Vercel AI SDK.

OpenAI Drop-in Replacement

The OpenAICompat export provides an OpenAI-compatible interface. Replace your OpenAI client with ClawPipe and get all pipeline optimizations automatically.

import { OpenAICompat } from 'clawpipe-ai';

// Drop-in replacement for OpenAI client
const openai = new OpenAICompat({
  apiKey: 'cp_xxx',
  projectId: 'my-app',
});

// Same API as OpenAI SDK
const completion = await openai.chat.completions.create({
  model: 'gpt-4o',
  messages: [
    { role: 'system', content: 'You are helpful.' },
    { role: 'user', content: 'Explain recursion' },
  ],
  max_tokens: 2000,
});

console.log(completion.choices[0].message.content);

LangChain

Use ClawPipe as a custom LLM provider in LangChain. Point LangChain's OpenAI integration at the ClawPipe gateway.

from langchain_openai import ChatOpenAI

# Point LangChain at ClawPipe gateway
llm = ChatOpenAI(
    base_url="https://api.clawpipe.ai/v1",
    api_key="cp_xxx",
    model="gpt-4o",
    default_headers={"X-Project-Id": "my-app"},
)

# Use exactly like normal LangChain
response = llm.invoke("Explain recursion")
print(response.content)

# Works with chains, agents, RAG, etc.
from langchain_core.prompts import ChatPromptTemplate

prompt = ChatPromptTemplate.from_messages([
    ("system", "You are a {role}."),
    ("human", "{input}"),
])
chain = prompt | llm
result = chain.invoke({"role": "tutor", "input": "What is a monad?"})

LlamaIndex

Configure LlamaIndex to use the ClawPipe gateway as its OpenAI-compatible backend.

from llama_index.llms.openai import OpenAI

# Use ClawPipe as the LLM backend
llm = OpenAI(
    api_base="https://api.clawpipe.ai/v1",
    api_key="cp_xxx",
    model="gpt-4o",
    additional_kwargs={"headers": {"X-Project-Id": "my-app"}},
)

# Use with any LlamaIndex component
from llama_index.core import VectorStoreIndex, SimpleDirectoryReader

documents = SimpleDirectoryReader("data").load_data()
index = VectorStoreIndex.from_documents(documents, llm=llm)
query_engine = index.as_query_engine()
response = query_engine.query("What does this codebase do?")

Vercel AI SDK

Use ClawPipe with the Vercel AI SDK by configuring the OpenAI provider with the ClawPipe gateway URL.

import { createOpenAI } from '@ai-sdk/openai';
import { generateText, streamText } from 'ai';

const clawpipe = createOpenAI({
  baseURL: 'https://api.clawpipe.ai/v1',
  apiKey: 'cp_xxx',
  headers: { 'X-Project-Id': 'my-app' },
});

// Generate text
const { text } = await generateText({
  model: clawpipe('gpt-4o'),
  prompt: 'Explain recursion',
});

// Stream text
const result = await streamText({
  model: clawpipe('gpt-4o'),
  prompt: 'Write a haiku about code',
});

for await (const chunk of result.textStream) {
  process.stdout.write(chunk);
}

Gateway as Proxy

Any OpenAI-compatible client can use ClawPipe by setting the base URL. This works with any language or framework that supports custom OpenAI endpoints.

# cURL example
curl -X POST https://api.clawpipe.ai/v1/prompt \
  -H "Authorization: Bearer cp_xxx" \
  -H "X-Project-Id: my-app" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "Explain recursion",
    "provider": "openai",
    "model": "gpt-4o",
    "maxTokens": 2000
  }'