langserve

LangServe 🦜️🏓

Stars: 1874

Visit

LangServe helps developers deploy `LangChain` runnables and chains as a REST API. This library is integrated with FastAPI and uses pydantic for data validation. In addition, it provides a client that can be used to call into runnables deployed on a server. A JavaScript client is available in LangChain.js.

README:

🦜️🏓 LangServe

Overview

LangServe helps developers deploy LangChain runnables and chains as a REST API.

This library is integrated with FastAPI and uses pydantic for data validation.

In addition, it provides a client that can be used to call into runnables deployed on a server. A JavaScript client is available in LangChain.js.

Features

Input and Output schemas automatically inferred from your LangChain object, and enforced on every API call, with rich error messages
API docs page with JSONSchema and Swagger (insert example link)
Efficient /invoke, /batch and /stream endpoints with support for many concurrent requests on a single server
/stream_log endpoint for streaming all (or some) intermediate steps from your chain/agent
new as of 0.0.40, supports /stream_events to make it easier to stream without needing to parse the output of /stream_log.
Playground page at /playground/ with streaming output and intermediate steps
Built-in (optional) tracing to LangSmith, just add your API key (see Instructions)
All built with battle-tested open-source Python libraries like FastAPI, Pydantic, uvloop and asyncio.
Use the client SDK to call a LangServe server as if it was a Runnable running locally (or call the HTTP API directly)
LangServe Hub

⚠️ LangGraph Compatibility

LangServe is designed to primarily deploy simple Runnables and wok with well-known primitives in langchain-core.

If you need a deployment option for LangGraph, you should instead be looking at LangGraph Cloud (beta) which will be better suited for deploying LangGraph applications.

Limitations

Client callbacks are not yet supported for events that originate on the server
Versions of LangServe <= 0.2.0, will not generate OpenAPI docs properly when using Pydantic V2 as Fast API does not support mixing pydantic v1 and v2 namespaces. See section below for more details. Either upgrade to LangServe>=0.3.0 or downgrade Pydantic to pydantic 1.

Security

Vulnerability in Versions 0.0.13 - 0.0.15 -- playground endpoint allows accessing arbitrary files on server. Resolved in 0.0.16.

Installation

For both client and server:

pip install "langserve[all]"

or pip install "langserve[client]" for client code, and pip install "langserve[server]" for server code.

LangChain CLI 🛠️

Use the LangChain CLI to bootstrap a LangServe project quickly.

To use the langchain CLI make sure that you have a recent version of langchain-cli installed. You can install it with pip install -U langchain-cli.

Setup

Note: We use poetry for dependency management. Please follow poetry doc to learn more about it.

1. Create new app using langchain cli command

langchain app new my-app

2. Define the runnable in add_routes. Go to server.py and edit

add_routes(app. NotImplemented)

3. Use `poetry` to add 3rd party packages (e.g., langchain-openai, langchain-anthropic, langchain-mistral etc).

poetry add [package-name] // e.g `poetry add langchain-openai`

4. Set up relevant env variables. For example,

export OPENAI_API_KEY="sk-..."

5. Serve your app

poetry run langchain serve --port=8100

Examples

Get your LangServe instance started quickly with LangChain Templates.

For more examples, see the templates index or the examples directory.

Description	Links
LLMs Minimal example that reserves OpenAI and Anthropic chat models. Uses async, supports batching and streaming.	server, client
Retriever Simple server that exposes a retriever as a runnable.	server, client
Conversational Retriever A Conversational Retriever exposed via LangServe	server, client
Agent without conversation history based on OpenAI tools	server, client
Agent with conversation history based on OpenAI tools	server, client
RunnableWithMessageHistory to implement chat persisted on backend, keyed off a `session_id` supplied by client.	server, client
RunnableWithMessageHistory to implement chat persisted on backend, keyed off a `conversation_id` supplied by client, and `user_id` (see Auth for implementing `user_id` properly).	server, client
Configurable Runnable to create a retriever that supports run time configuration of the index name.	server, client
Configurable Runnable that shows configurable fields and configurable alternatives.	server, client
APIHandler Shows how to use `APIHandler` instead of `add_routes`. This provides more flexibility for developers to define endpoints. Works well with all FastAPI patterns, but takes a bit more effort.	server
LCEL Example Example that uses LCEL to manipulate a dictionary input.	server, client
Auth with `add_routes`: Simple authentication that can be applied across all endpoints associated with app. (Not useful on its own for implementing per user logic.)	server
Auth with `add_routes`: Simple authentication mechanism based on path dependencies. (No useful on its own for implementing per user logic.)	server
Auth with `add_routes`: Implement per user logic and auth for endpoints that use per request config modifier. (Note: At the moment, does not integrate with OpenAPI docs.)	server, client
Auth with `APIHandler`: Implement per user logic and auth that shows how to search only within user owned documents.	server, client
Widgets Different widgets that can be used with playground (file upload and chat)	server
Widgets File upload widget used for LangServe playground.	server, client

Sample Application

Server

Here's a server that deploys an OpenAI chat model, an Anthropic chat model, and a chain that uses the Anthropic model to tell a joke about a topic.

#!/usr/bin/env python
from fastapi import FastAPI
from langchain.prompts import ChatPromptTemplate
from langchain.chat_models import ChatAnthropic, ChatOpenAI
from langserve import add_routes

app = FastAPI(
    title="LangChain Server",
    version="1.0",
    description="A simple api server using Langchain's Runnable interfaces",
)

add_routes(
    app,
    ChatOpenAI(model="gpt-3.5-turbo-0125"),
    path="/openai",
)

add_routes(
    app,
    ChatAnthropic(model="claude-3-haiku-20240307"),
    path="/anthropic",
)

model = ChatAnthropic(model="claude-3-haiku-20240307")
prompt = ChatPromptTemplate.from_template("tell me a joke about {topic}")
add_routes(
    app,
    prompt | model,
    path="/joke",
)

if __name__ == "__main__":
    import uvicorn

    uvicorn.run(app, host="localhost", port=8000)

If you intend to call your endpoint from the browser, you will also need to set CORS headers. You can use FastAPI's built-in middleware for that:

from fastapi.middleware.cors import CORSMiddleware

# Set all CORS enabled origins
app.add_middleware(
    CORSMiddleware,
    allow_origins=["*"],
    allow_credentials=True,
    allow_methods=["*"],
    allow_headers=["*"],
    expose_headers=["*"],
)

Docs

If you've deployed the server above, you can view the generated OpenAPI docs using:

⚠️ If using LangServe <= 0.2.0 and pydantic v2, docs will not be generated for invoke, batch, stream, stream_log. See Pydantic section below for more details. To resolve please upgrade to LangServe 0.3.0.

curl localhost:8000/docs

make sure to add the /docs suffix.

⚠️ Index page / is not defined by design, so curl localhost:8000 or visiting the URL will return a 404. If you want content at / define an endpoint @app.get("/").

Client

Python SDK

from langchain.schema import SystemMessage, HumanMessage
from langchain.prompts import ChatPromptTemplate
from langchain.schema.runnable import RunnableMap
from langserve import RemoteRunnable

openai = RemoteRunnable("http://localhost:8000/openai/")
anthropic = RemoteRunnable("http://localhost:8000/anthropic/")
joke_chain = RemoteRunnable("http://localhost:8000/joke/")

joke_chain.invoke({"topic": "parrots"})

# or async
await joke_chain.ainvoke({"topic": "parrots"})

prompt = [
    SystemMessage(content='Act like either a cat or a parrot.'),
    HumanMessage(content='Hello!')
]

# Supports astream
async for msg in anthropic.astream(prompt):
    print(msg, end="", flush=True)

prompt = ChatPromptTemplate.from_messages(
    [("system", "Tell me a long story about {topic}")]
)

# Can define custom chains
chain = prompt | RunnableMap({
    "openai": openai,
    "anthropic": anthropic,
})

chain.batch([{"topic": "parrots"}, {"topic": "cats"}])

In TypeScript (requires LangChain.js version 0.0.166 or later):

import { RemoteRunnable } from "@langchain/core/runnables/remote";

const chain = new RemoteRunnable({
  url: `http://localhost:8000/joke/`,
});
const result = await chain.invoke({
  topic: "cats",
});

Python using requests:

import requests

response = requests.post(
    "http://localhost:8000/joke/invoke",
    json={'input': {'topic': 'cats'}}
)
response.json()

You can also use curl:

curl --location --request POST 'http://localhost:8000/joke/invoke' \
    --header 'Content-Type: application/json' \
    --data-raw '{
        "input": {
            "topic": "cats"
        }
    }'

Endpoints

The following code:

...
add_routes(
    app,
    runnable,
    path="/my_runnable",
)

adds of these endpoints to the server:

POST /my_runnable/invoke - invoke the runnable on a single input
POST /my_runnable/batch - invoke the runnable on a batch of inputs
POST /my_runnable/stream - invoke on a single input and stream the output
POST /my_runnable/stream_log - invoke on a single input and stream the output, including output of intermediate steps as it's generated
POST /my_runnable/astream_events - invoke on a single input and stream events as they are generated, including from intermediate steps.
GET /my_runnable/input_schema - json schema for input to the runnable
GET /my_runnable/output_schema - json schema for output of the runnable
GET /my_runnable/config_schema - json schema for config of the runnable

These endpoints match the LangChain Expression Language interface -- please reference this documentation for more details.

Playground

You can find a playground page for your runnable at /my_runnable/playground/. This exposes a simple UI to configure and invoke your runnable with streaming output and intermediate steps.

Widgets

The playground supports widgets and can be used to test your runnable with different inputs. See the widgets section below for more details.

Sharing

In addition, for configurable runnables, the playground will allow you to configure the runnable and share a link with the configuration:

Chat playground

LangServe also supports a chat-focused playground that opt into and use under /my_runnable/playground/. Unlike the general playground, only certain types of runnables are supported - the runnable's input schema must be a dict with either:

a single key, and that key's value must be a list of chat messages.
two keys, one whose value is a list of messages, and the other representing the most recent message.

We recommend you use the first format.

The runnable must also return either an AIMessage or a string.

To enable it, you must set playground_type="chat", when adding your route. Here's an example:

# Declare a chain
prompt = ChatPromptTemplate.from_messages(
    [
        ("system", "You are a helpful, professional assistant named Cob."),
        MessagesPlaceholder(variable_name="messages"),
    ]
)

chain = prompt | ChatAnthropic(model="claude-2.1")


class InputChat(BaseModel):
    """Input for the chat endpoint."""

    messages: List[Union[HumanMessage, AIMessage, SystemMessage]] = Field(
        ...,
        description="The chat messages representing the current conversation.",
    )


add_routes(
    app,
    chain.with_types(input_type=InputChat),
    enable_feedback_endpoint=True,
    enable_public_trace_link_endpoint=True,
    playground_type="chat",
)

If you are using LangSmith, you can also set enable_feedback_endpoint=True on your route to enable thumbs-up/thumbs-down buttons after each message, and enable_public_trace_link_endpoint=True to add a button that creates a public traces for runs. Note that you will also need to set the following environment variables:

export LANGCHAIN_TRACING_V2="true"
export LANGCHAIN_PROJECT="YOUR_PROJECT_NAME"
export LANGCHAIN_API_KEY="YOUR_API_KEY"

Here's an example with the above two options turned on:

Note: If you enable public trace links, the internals of your chain will be exposed. We recommend only using this setting for demos or testing.

Legacy Chains

LangServe works with both Runnables (constructed via LangChain Expression Language) and legacy chains (inheriting from Chain). However, some of the input schemas for legacy chains may be incomplete/incorrect, leading to errors. This can be fixed by updating the input_schema property of those chains in LangChain. If you encounter any errors, please open an issue on THIS repo, and we will work to address it.

Deployment

Deploy to AWS

You can deploy to AWS using the AWS Copilot CLI

copilot init --app [application-name] --name [service-name] --type 'Load Balanced Web Service' --dockerfile './Dockerfile' --deploy

Click here to learn more.

Deploy to Azure

You can deploy to Azure using Azure Container Apps (Serverless):

az containerapp up --name [container-app-name] --source . --resource-group [resource-group-name] --environment  [environment-name] --ingress external --target-port 8001 --env-vars=OPENAI_API_KEY=your_key

You can find more info here

Deploy to GCP

You can deploy to GCP Cloud Run using the following command:

gcloud run deploy [your-service-name] --source . --port 8001 --allow-unauthenticated --region us-central1 --set-env-vars=OPENAI_API_KEY=your_key

Community Contributed

Deploy to Railway

Example Railway Repo

Pydantic

LangServe provides support for Pydantic 2 with some limitations.

OpenAPI docs will not be generated for invoke/batch/stream/stream_log when using Pydantic V2. Fast API does not support [mixing pydantic v1 and v2 namespaces]. To fix this, use pip install pydantic==1.10.17.
LangChain uses the v1 namespace in Pydantic v2. Please read the following guidelines to ensure compatibility with LangChain

Except for these limitations, we expect the API endpoints, the playground and any other features to work as expected.

Advanced

Handling Authentication

If you need to add authentication to your server, please read Fast API's documentation about dependencies and security.

The below examples show how to wire up authentication logic LangServe endpoints using FastAPI primitives.

You are responsible for providing the actual authentication logic, the users table etc.

If you're not sure what you're doing, you could try using an existing solution Auth0.

Using add_routes

If you're using add_routes, see examples here.

Description	Links
Auth with `add_routes`: Simple authentication that can be applied across all endpoints associated with app. (Not useful on its own for implementing per user logic.)	server
Auth with `add_routes`: Simple authentication mechanism based on path dependencies. (No useful on its own for implementing per user logic.)	server
Auth with `add_routes`: Implement per user logic and auth for endpoints that use per request config modifier. (Note: At the moment, does not integrate with OpenAPI docs.)	server, client

Alternatively, you can use FastAPI's middleware.

Using global dependencies and path dependencies has the advantage that auth will be properly supported in the OpenAPI docs page, but these are not sufficient for implement per user logic (e.g., making an application that can search only within user owned documents).

If you need to implement per user logic, you can use the per_req_config_modifier or APIHandler (below) to implement this logic.

Per User

If you need authorization or logic that is user dependent, specify per_req_config_modifier when using add_routes. Use a callable receives the raw Request object and can extract relevant information from it for authentication and authorization purposes.

Using APIHandler

If you feel comfortable with FastAPI and python, you can use LangServe's APIHandler.

Description	Links
Auth with `APIHandler`: Implement per user logic and auth that shows how to search only within user owned documents.	server, client
APIHandler Shows how to use `APIHandler` instead of `add_routes`. This provides more flexibility for developers to define endpoints. Works well with all FastAPI patterns, but takes a bit more effort.	server, client

It's a bit more work, but gives you complete control over the endpoint definitions, so you can do whatever custom logic you need for auth.

Files

LLM applications often deal with files. There are different architectures that can be made to implement file processing; at a high level:

The file may be uploaded to the server via a dedicated endpoint and processed using a separate endpoint
The file may be uploaded by either value (bytes of file) or reference (e.g., s3 url to file content)
The processing endpoint may be blocking or non-blocking
If significant processing is required, the processing may be offloaded to a dedicated process pool

You should determine what is the appropriate architecture for your application.

Currently, to upload files by value to a runnable, use base64 encoding for the file (multipart/form-data is not supported yet).

Here's an example that shows how to use base64 encoding to send a file to a remote runnable.

Remember, you can always upload files by reference (e.g., s3 url) or upload them as multipart/form-data to a dedicated endpoint.

Custom Input and Output Types

Input and Output types are defined on all runnables.

You can access them via the input_schema and output_schema properties.

LangServe uses these types for validation and documentation.

If you want to override the default inferred types, you can use the with_types method.

Here's a toy example to illustrate the idea:

from typing import Any

from fastapi import FastAPI
from langchain.schema.runnable import RunnableLambda

app = FastAPI()


def func(x: Any) -> int:
    """Mistyped function that should accept an int but accepts anything."""
    return x + 1


runnable = RunnableLambda(func).with_types(
    input_type=int,
)

add_routes(app, runnable)

Custom User Types

Inherit from CustomUserType if you want the data to de-serialize into a pydantic model rather than the equivalent dict representation.

At the moment, this type only works server side and is used to specify desired decoding behavior. If inheriting from this type the server will keep the decoded type as a pydantic model instead of converting it into a dict.

from fastapi import FastAPI
from langchain.schema.runnable import RunnableLambda

from langserve import add_routes
from langserve.schema import CustomUserType

app = FastAPI()


class Foo(CustomUserType):
    bar: int


def func(foo: Foo) -> int:
    """Sample function that expects a Foo type which is a pydantic model"""
    assert isinstance(foo, Foo)
    return foo.bar


# Note that the input and output type are automatically inferred!
# You do not need to specify them.
# runnable = RunnableLambda(func).with_types( # <-- Not needed in this case
#     input_type=Foo,
#     output_type=int,
#
add_routes(app, RunnableLambda(func), path="/foo")

Playground Widgets

The playground allows you to define custom widgets for your runnable from the backend.

Here are a few examples:

Description	Links
Widgets Different widgets that can be used with playground (file upload and chat)	server, client
Widgets File upload widget used for LangServe playground.	server, client

Schema

A widget is specified at the field level and shipped as part of the JSON schema of the input type
A widget must contain a key called type with the value being one of a well known list of widgets
Other widget keys will be associated with values that describe paths in a JSON object

type JsonPath = number | string | (number | string)[];
type NameSpacedPath = { title: string; path: JsonPath }; // Using title to mimick json schema, but can use namespace
type OneOfPath = { oneOf: JsonPath[] };

type Widget = {
  type: string; // Some well known type (e.g., base64file, chat etc.)
  [key: string]: JsonPath | NameSpacedPath | OneOfPath;
};

Available Widgets

There are only two widgets that the user can specify manually right now:

File Upload Widget
Chat History Widget

See below more information about these widgets.

All other widgets on the playground UI are created and managed automatically by the UI based on the config schema of the Runnable. When you create Configurable Runnables, the playground should create appropriate widgets for you to control the behavior.

File Upload Widget

Allows creation of a file upload input in the UI playground for files that are uploaded as base64 encoded strings. Here's the full example.

Snippet:

try:
    from pydantic.v1 import Field
except ImportError:
    from pydantic import Field

from langserve import CustomUserType


# ATTENTION: Inherit from CustomUserType instead of BaseModel otherwise
#            the server will decode it into a dict instead of a pydantic model.
class FileProcessingRequest(CustomUserType):
    """Request including a base64 encoded file."""

    # The extra field is used to specify a widget for the playground UI.
    file: str = Field(..., extra={"widget": {"type": "base64file"}})
    num_chars: int = 100

Example widget:

Chat Widget

Look at the widget example.

To define a chat widget, make sure that you pass "type": "chat".

"input" is JSONPath to the field in the Request that has the new input message.
"output" is JSONPath to the field in the Response that has new output message(s).
Don't specify these fields if the entire input or output should be used as they are ( e.g., if the output is a list of chat messages.)

Here's a snippet:

class ChatHistory(CustomUserType):
    chat_history: List[Tuple[str, str]] = Field(
        ...,
        examples=[[("human input", "ai response")]],
        extra={"widget": {"type": "chat", "input": "question", "output": "answer"}},
    )
    question: str


def _format_to_messages(input: ChatHistory) -> List[BaseMessage]:
    """Format the input to a list of messages."""
    history = input.chat_history
    user_input = input.question

    messages = []

    for human, ai in history:
        messages.append(HumanMessage(content=human))
        messages.append(AIMessage(content=ai))
    messages.append(HumanMessage(content=user_input))
    return messages


model = ChatOpenAI()
chat_model = RunnableParallel({"answer": (RunnableLambda(_format_to_messages) | model)})
add_routes(
    app,
    chat_model.with_types(input_type=ChatHistory),
    config_keys=["configurable"],
    path="/chat",
)

Example widget:

You can also specify a list of messages as your a parameter directly, as shown in this snippet:

prompt = ChatPromptTemplate.from_messages(
    [
        ("system", "You are a helpful assisstant named Cob."),
        MessagesPlaceholder(variable_name="messages"),
    ]
)

chain = prompt | ChatAnthropic(model="claude-2.1")


class MessageListInput(BaseModel):
    """Input for the chat endpoint."""
    messages: List[Union[HumanMessage, AIMessage]] = Field(
        ...,
        description="The chat messages representing the current conversation.",
        extra={"widget": {"type": "chat", "input": "messages"}},
    )


add_routes(
    app,
    chain.with_types(input_type=MessageListInput),
    path="/chat",
)

See this sample file for an example.

Enabling / Disabling Endpoints (LangServe >=0.0.33)

You can enable / disable which endpoints are exposed when adding routes for a given chain.

Use enabled_endpoints if you want to make sure to never get a new endpoint when upgrading langserve to a newer verison.

Enable: The code below will only enable invoke, batch and the corresponding config_hash endpoint variants.

add_routes(app, chain, enabled_endpoints=["invoke", "batch", "config_hashes"], path="/mychain")

Disable: The code below will disable the playground for the chain

add_routes(app, chain, disabled_endpoints=["playground"], path="/mychain")

For Tasks:

Click tags to check more tools for each tasks

invoke batch stream stream_log stream_events

For Jobs:

chatbot content generation data analysis image processing translation

Alternative AI tools for langserve

Similar Open Source Tools

langserve

github

: 1.9k

client-python

The Mistral Python Client is a tool inspired by cohere-python that allows users to interact with the Mistral AI API. It provides functionalities to access and utilize the AI capabilities offered by Mistral. Users can easily install the client using pip and manage dependencies using poetry. The client includes examples demonstrating how to use the API for various tasks, such as chat interactions. To get started, users need to obtain a Mistral API Key and set it as an environment variable. Overall, the Mistral Python Client simplifies the integration of Mistral AI services into Python applications.

github

: 451

upgini

Upgini is an intelligent data search engine with a Python library that helps users find and add relevant features to their ML pipeline from various public, community, and premium external data sources. It automates the optimization of connected data sources by generating an optimal set of machine learning features using large language models, GraphNNs, and recurrent neural networks. The tool aims to simplify feature search and enrichment for external data to make it a standard approach in machine learning pipelines. It democratizes access to data sources for the data science community.

github

: 314

llm-vscode

llm-vscode is an extension designed for all things LLM, utilizing llm-ls as its backend. It offers features such as code completion with 'ghost-text' suggestions, the ability to choose models for code generation via HTTP requests, ensuring prompt size fits within the context window, and code attribution checks. Users can configure the backend, suggestion behavior, keybindings, llm-ls settings, and tokenization options. Additionally, the extension supports testing models like Code Llama 13B, Phind/Phind-CodeLlama-34B-v2, and WizardLM/WizardCoder-Python-34B-V1.0. Development involves cloning llm-ls, building it, and setting up the llm-vscode extension for use.

github

: 1.1k

syncode

SynCode is a novel framework for the grammar-guided generation of Large Language Models (LLMs) that ensures syntactically valid output with respect to defined Context-Free Grammar (CFG) rules. It supports general-purpose programming languages like Python, Go, SQL, JSON, and more, allowing users to define custom grammars using EBNF syntax. The tool compares favorably to other constrained decoders and offers features like fast grammar-guided generation, compatibility with HuggingFace Language Models, and the ability to work with various decoding strategies.

github

: 165

hydraai

Generate React components on-the-fly at runtime using AI. Register your components, and let Hydra choose when to show them in your App. Hydra development is still early, and patterns for different types of components and apps are still being developed. Join the discord to chat with the developers. Expects to be used in a NextJS project. Components that have function props do not work.

github

: 90

Gemini-API

Gemini-API is a reverse-engineered asynchronous Python wrapper for Google Gemini web app (formerly Bard). It provides features like persistent cookies, ImageFx support, extension support, classified outputs, official flavor, and asynchronous operation. The tool allows users to generate contents from text or images, have conversations across multiple turns, retrieve images in response, generate images with ImageFx, save images to local files, use Gemini extensions, check and switch reply candidates, and control log level.

github

: 160

mistral-inference

Mistral Inference repository contains minimal code to run 7B, 8x7B, and 8x22B models. It provides model download links, installation instructions, and usage guidelines for running models via CLI or Python. The repository also includes information on guardrailing, model platforms, deployment, and references. Users can interact with models through commands like mistral-demo, mistral-chat, and mistral-common. Mistral AI models support function calling and chat interactions for tasks like testing models, chatting with models, and using Codestral as a coding assistant. The repository offers detailed documentation and links to blogs for further information.

github

: 9.6k

lollms

LoLLMs Server is a text generation server based on large language models. It provides a Flask-based API for generating text using various pre-trained language models. This server is designed to be easy to install and use, allowing developers to integrate powerful text generation capabilities into their applications.

github

: 266

IntelliNode

IntelliNode is a javascript module that integrates cutting-edge AI models like ChatGPT, LLaMA, WaveNet, Gemini, and Stable diffusion into projects. It offers functions for generating text, speech, and images, as well as semantic search, multi-model evaluation, and chatbot capabilities. The module provides a wrapper layer for low-level model access, a controller layer for unified input handling, and a function layer for abstract functionality tailored to various use cases.

github

: 201

llmgraph

llmgraph is a tool that enables users to create knowledge graphs in GraphML, GEXF, and HTML formats by extracting world knowledge from large language models (LLMs) like ChatGPT. It supports various entity types and relationships, offers cache support for efficient graph growth, and provides insights into LLM costs. Users can customize the model used and interact with different LLM providers. The tool allows users to generate interactive graphs based on a specified entity type and Wikipedia link, making it a valuable resource for knowledge graph creation and exploration.

github

: 271

call-center-ai

Call Center AI is an AI-powered call center solution that leverages Azure and OpenAI GPT. It is a proof of concept demonstrating the integration of Azure Communication Services, Azure Cognitive Services, and Azure OpenAI to build an automated call center solution. The project showcases features like accessing claims on a public website, customer conversation history, language change during conversation, bot interaction via phone number, multiple voice tones, lexicon understanding, todo list creation, customizable prompts, content filtering, GPT-4 Turbo for customer requests, specific data schema for claims, documentation database access, SMS report sending, conversation resumption, and more. The system architecture includes components like RAG AI Search, SMS gateway, call gateway, moderation, Cosmos DB, event broker, GPT-4 Turbo, Redis cache, translation service, and more. The tool can be deployed remotely using GitHub Actions and locally with prerequisites like Azure environment setup, configuration file creation, and resource hosting. Advanced usage includes custom training data with AI Search, prompt customization, language customization, moderation level customization, claim data schema customization, OpenAI compatible model usage for the LLM, and Twilio integration for SMS.

github

: 119

Lumos

Lumos is a Chrome extension powered by a local LLM co-pilot for browsing the web. It allows users to summarize long threads, news articles, and technical documentation. Users can ask questions about reviews and product pages. The tool requires a local Ollama server for LLM inference and embedding database. Lumos supports multimodal models and file attachments for processing text and image content. It also provides options to customize models, hosts, and content parsers. The extension can be easily accessed through keyboard shortcuts and offers tools for automatic invocation based on prompts.

github

: 1.3k

HippoRAG

HippoRAG is a novel retrieval augmented generation (RAG) framework inspired by the neurobiology of human long-term memory that enables Large Language Models (LLMs) to continuously integrate knowledge across external documents. It provides RAG systems with capabilities that usually require a costly and high-latency iterative LLM pipeline for only a fraction of the computational cost. The tool facilitates setting up retrieval corpus, indexing, and retrieval processes for LLMs, offering flexibility in choosing different online LLM APIs or offline LLM deployments through LangChain integration. Users can run retrieval on pre-defined queries or integrate directly with the HippoRAG API. The tool also supports reproducibility of experiments and provides data, baselines, and hyperparameter tuning scripts for research purposes.

github

: 1.1k

exif-photo-blog

EXIF Photo Blog is a full-stack photo blog application built with Next.js, Vercel, and Postgres. It features built-in authentication, photo upload with EXIF extraction, photo organization by tag, infinite scroll, light/dark mode, automatic OG image generation, a CMD-K menu with photo search, experimental support for AI-generated descriptions, and support for Fujifilm simulations. The application is easy to deploy to Vercel with just a few clicks and can be customized with a variety of environment variables.

github

: 609

chatgpt-cli

ChatGPT CLI provides a powerful command-line interface for seamless interaction with ChatGPT models via OpenAI and Azure. It features streaming capabilities, extensive configuration options, and supports various modes like streaming, query, and interactive mode. Users can manage thread-based context, sliding window history, and provide custom context from any source. The CLI also offers model and thread listing, advanced configuration options, and supports GPT-4, GPT-3.5-turbo, and Perplexity's models. Installation is available via Homebrew or direct download, and users can configure settings through default values, a config.yaml file, or environment variables.

github

: 519

For similar tasks

langserve

github

: 1.9k

For similar jobs

h2ogpt

h2oGPT is an Apache V2 open-source project that allows users to query and summarize documents or chat with local private GPT LLMs. It features a private offline database of any documents (PDFs, Excel, Word, Images, Video Frames, Youtube, Audio, Code, Text, MarkDown, etc.), a persistent database (Chroma, Weaviate, or in-memory FAISS) using accurate embeddings (instructor-large, all-MiniLM-L6-v2, etc.), and efficient use of context using instruct-tuned LLMs (no need for LangChain's few-shot approach). h2oGPT also offers parallel summarization and extraction, reaching an output of 80 tokens per second with the 13B LLaMa2 model, HYDE (Hypothetical Document Embeddings) for enhanced retrieval based upon LLM responses, a variety of models supported (LLaMa2, Mistral, Falcon, Vicuna, WizardLM. With AutoGPTQ, 4-bit/8-bit, LORA, etc.), GPU support from HF and LLaMa.cpp GGML models, and CPU support using HF, LLaMa.cpp, and GPT4ALL models. Additionally, h2oGPT provides Attention Sinks for arbitrarily long generation (LLaMa-2, Mistral, MPT, Pythia, Falcon, etc.), a UI or CLI with streaming of all models, the ability to upload and view documents through the UI (control multiple collaborative or personal collections), Vision Models LLaVa, Claude-3, Gemini-Pro-Vision, GPT-4-Vision, Image Generation Stable Diffusion (sdxl-turbo, sdxl) and PlaygroundAI (playv2), Voice STT using Whisper with streaming audio conversion, Voice TTS using MIT-Licensed Microsoft Speech T5 with multiple voices and Streaming audio conversion, Voice TTS using MPL2-Licensed TTS including Voice Cloning and Streaming audio conversion, AI Assistant Voice Control Mode for hands-free control of h2oGPT chat, Bake-off UI mode against many models at the same time, Easy Download of model artifacts and control over models like LLaMa.cpp through the UI, Authentication in the UI by user/password via Native or Google OAuth, State Preservation in the UI by user/password, Linux, Docker, macOS, and Windows support, Easy Windows Installer for Windows 10 64-bit (CPU/CUDA), Easy macOS Installer for macOS (CPU/M1/M2), Inference Servers support (oLLaMa, HF TGI server, vLLM, Gradio, ExLLaMa, Replicate, OpenAI, Azure OpenAI, Anthropic), OpenAI-compliant, Server Proxy API (h2oGPT acts as drop-in-replacement to OpenAI server), Python client API (to talk to Gradio server), JSON Mode with any model via code block extraction. Also supports MistralAI JSON mode, Claude-3 via function calling with strict Schema, OpenAI via JSON mode, and vLLM via guided_json with strict Schema, Web-Search integration with Chat and Document Q/A, Agents for Search, Document Q/A, Python Code, CSV frames (Experimental, best with OpenAI currently), Evaluate performance using reward models, and Quality maintained with over 1000 unit and integration tests taking over 4 GPU-hours.

github

: 11.3k

mistral.rs

Mistral.rs is a fast LLM inference platform written in Rust. We support inference on a variety of devices, quantization, and easy-to-use application with an Open-AI API compatible HTTP server and Python bindings.

github

: 3.7k

ollama

Ollama is a lightweight, extensible framework for building and running language models on the local machine. It provides a simple API for creating, running, and managing models, as well as a library of pre-built models that can be easily used in a variety of applications. Ollama is designed to be easy to use and accessible to developers of all levels. It is open source and available for free on GitHub.

github

: 92.5k

llama-cpp-agent

The llama-cpp-agent framework is a tool designed for easy interaction with Large Language Models (LLMs). Allowing users to chat with LLM models, execute structured function calls and get structured output (objects). It provides a simple yet robust interface and supports llama-cpp-python and OpenAI endpoints with GBNF grammar support (like the llama-cpp-python server) and the llama.cpp backend server. It works by generating a formal GGML-BNF grammar of the user defined structures and functions, which is then used by llama.cpp to generate text valid to that grammar. In contrast to most GBNF grammar generators it also supports nested objects, dictionaries, enums and lists of them.

github

: 454

llama_ros

This repository provides a set of ROS 2 packages to integrate llama.cpp into ROS 2. By using the llama_ros packages, you can easily incorporate the powerful optimization capabilities of llama.cpp into your ROS 2 projects by running GGUF-based LLMs and VLMs.

github

: 146

MITSUHA

OneReality is a virtual waifu/assistant that you can speak to through your mic and it'll speak back to you! It has many features such as: * You can speak to her with a mic * It can speak back to you * Has short-term memory and long-term memory * Can open apps * Smarter than you * Fluent in English, Japanese, Korean, and Chinese * Can control your smart home like Alexa if you set up Tuya (more info in Prerequisites) It is built with Python, Llama-cpp-python, Whisper, SpeechRecognition, PocketSphinx, VITS-fast-fine-tuning, VITS-simple-api, HyperDB, Sentence Transformers, and Tuya Cloud IoT.

github

: 174

wenxin-starter

WenXin-Starter is a spring-boot-starter for Baidu's "Wenxin Qianfan WENXINWORKSHOP" large model, which can help you quickly access Baidu's AI capabilities. It fully integrates the official API documentation of Wenxin Qianfan. Supports text-to-image generation, built-in dialogue memory, and supports streaming return of dialogue. Supports QPS control of a single model and supports queuing mechanism. Plugins will be added soon.

github

: 207

FlexFlow

FlexFlow Serve is an open-source compiler and distributed system for **low latency**, **high performance** LLM serving. FlexFlow Serve outperforms existing systems by 1.3-2.0x for single-node, multi-GPU inference and by 1.4-2.4x for multi-node, multi-GPU inference.

github

: 1.7k

langserve

README:

🦜️🏓 LangServe

Overview

Features

⚠️ LangGraph Compatibility

Limitations

Security

Installation

LangChain CLI 🛠️

Setup

1. Create new app using langchain cli command

2. Define the runnable in add_routes. Go to server.py and edit

3. Use poetry to add 3rd party packages (e.g., langchain-openai, langchain-anthropic, langchain-mistral etc).

4. Set up relevant env variables. For example,

5. Serve your app

Examples

Sample Application

Server

Docs

Client

Endpoints

Playground

Widgets

Sharing

Chat playground

Legacy Chains

Deployment

Deploy to AWS

Deploy to Azure

Deploy to GCP

Community Contributed

Deploy to Railway

Pydantic

Advanced

Handling Authentication

Using add_routes

Using APIHandler

Files

Custom Input and Output Types

Custom User Types

Playground Widgets

Schema

Available Widgets

File Upload Widget

Chat Widget

Enabling / Disabling Endpoints (LangServe >=0.0.33)

For Tasks:

For Jobs:

Alternative AI tools for langserve

Similar Open Source Tools

langserve

client-python

upgini

llm-vscode

syncode

hydraai

Gemini-API

mistral-inference

lollms

IntelliNode

llmgraph

call-center-ai

Lumos

HippoRAG

exif-photo-blog

chatgpt-cli

For similar tasks

langserve

For similar jobs

h2ogpt

mistral.rs

ollama

llama-cpp-agent

llama_ros

MITSUHA

wenxin-starter

FlexFlow

3. Use `poetry` to add 3rd party packages (e.g., langchain-openai, langchain-anthropic, langchain-mistral etc).