Agent frameworks & infrastructure

Agent frameworks & SDKs

12,596 agents · most starred first

Agent frameworks & SDKs, page 85 of 126

llm-axe

A toolkit for quickly implementing llm powered functionalities.

llm-blender

LLM-Blender, an innovative ensembling framework to attain consistently superior performance by leveraging the diverse strengths and weaknesses of multiple open-source large language models (LLMs). LLM-Blender cut the weaknesses through ranking and integrate the strengths through fusing generation to enhance the capability of LLMs.

llm-blocks

Simple interface for creating and managing LLM chains

llm-bot

Python library for developing LLM bots

llm-chatbot-kit

Developer-first kit for LLM chatbots with streaming, memory, and personas (Discord-first)

llm-checks-common-functions

common functions, utilities using multiple DB and LLM options.

llm-classifier

Structured LLM based classification, clustering and extraction framework that works with all major API providers

llm-classifiers

Framework to build production-grade LLM-based classifiers

llm-classify

LLM plugin for content classification using various language models

llm-cleaner

A robust JSON repair and extraction tool for LLM outputs.

llm-client

SDK for using LLM

llm-clients

A reusable library for LLM clients with LangChain framework

llm-components

A library of ready-to-use LLM components

llm-conversation

A tool for LLM agent conversations

llm-data-processor

A processor for LLM tasks

llm-dataclass

A dataclass interface for llms

llm-director

A light-weight Python library 🐍 for deploying DAG chains of customizable Large Language Model operations. Leveraging the asyncio framework, LLM-Director allows developers to create, manage, and orchestrate actions and processes in an efficient and scalable manner. With its unique structure and integration with FastAPI, LLM-Director offes a robust solution for building applications, API endpoints, and workflow automation systems.

llm-dot-py

A modular, block-by-block LLM building library

llm-flow

A prompt graph definition and execution library for LLM app development.

llm-formats

Minimal llm_formats package exposing a single public function

llm-foundation

LLM Foundation Tools

llm-fsm

Finite State Machines for Large Language Models

llm-function

Llm function is a decorator that uses llm to assemble reusable workflows from available tools to match input and output models.

llm-function-tools

Minimal metadata and discovery helpers for typed Python tools used by llm_function runtimes.

llm-functional-agents

A lightweight Python library for creating reliable, contract-driven LLM functions whose core logic is implemented by an LLM.

llm-functions

Build semantic functions.

llm-gent

Agent framework with learning capabilities

llm-graph-kit

LLM ベースのエージェントのワークフローをグラフ形式で記述・実行する軽量ライブラリ

llm-harness

Lightweight AI agent infrastructure — harness orchestrates, backends execute

llm-helper-tools

LLM utilities and tools

llm-helpers

A helper package to work with LLMs

llm-interaction

Async multi-provider LLM client with typed tool calling, lazy output parsing, and Azure/Databricks/OpenRouter/LiteLLM/DeepSeek/local backends

llm-kit

A pythonic way to call llms & implement llm algorithms.

llm-latchkey

An LLM tool plugin for using Latchkey to make authenticated HTTP API requests.

llm-loop

A utility package for querying language models with pattern matching and retry logic

llm-magics

LLM Magics for Ipython created by kmkolasinski

llm-markdown

Turn Python functions into typed LLM calls using docstrings as prompts

llm-matrix

llm_matrix

llm-mcts-inference

An experimental project using Monte Carlo Tree Search (MCTS) to refine LLM responses for better accuracy and decision-making.

llm-mindmeld

A framework to glue all your AI tools together

llm-os

An open-source chatgpt tool ecosystem where you can combine tools with chatgpt and use natural language to do anything.

llm-patch-driver

LLM Patch Driver is a framework for patching data objects using LLMs. It can generate and apply a single patch, or start the patching loop to fix complex validation issues.

llm-pipelines

A modular, async, composable AI pipeline library for building conversational agents and content processing pipelines

llm-protocol-suite

Pydantic data models and adapters for the LLM Protocol Suite.

llm-queue-task-manager

A Redis-based queue task manager for LLM processing with intelligent token budget management

llm-reasoners

A library for advanced reasoning methods with large language models

llm-recursive

A Python framework for building intelligent, recursive task decomposition systems powered by Large Language Models

llm-reflection

This system utilizes a large language model (LLM) and reflection

llm-repeator-redis

LLM tools for repeated generation

llm-research

A minimum Python package built on top of the LangChain framework to interact with LLM.

llm-rpc

A Python package for LLM RPC operations

llm-rsa

Recursive self aggregation

llm-saia

Framework-agnostic verb vocabulary for LLM agents

llm-salvage

Salvage structured data from LLM responses that didn't follow instructions.

llm-sampler

Microlib for sampling from an LLM

llm-samplers

A library for advanced LLM sampling techniques

llm-schema-lite

LLM-ify your JSON schemas

llm-schema-lock

Fix broken JSON from Ollama, llama.cpp, vLLM. Auto-repair and offline validation.

llm-scratch

A collection of LLM utils

llm-sdk

This is a python sdk for llm

llm-sdk-core

Provider-agnostic SDK for LLMs with async support and plugin providers.

llm-sdk-py

Small, clean Python SDK for OpenAI-compatible LLM APIs.

llm-stack-kit

Generate Data from LLM easily using LLMKit

llm-station

Agent framework with tool calling for OpenAI, Anthropic, and Google

llm-stream-parser

Real-time parser for LLM streaming responses with tag-based content extraction

llm-strings

A brief description of my package

llm-tap

lightweight library for building Trigger-Action programs with llm

llm-task-handler

LLM Task Handler

llm-tasker

A lightweight Python library for creating tasks and workflows with Large Language Models (LLMs) using prompt templates.

llm-to-invoke
llm-tool
llm-tool-fusion

Biblioteca Python que simplifica e unifica a definição e chamada de ferramentas para grandes modelos de linguagem (LLMs). Compatível com Ollama, LangChain, OpenAI e outros frameworks.

llm-tool-harness

A library to easily build LLM agents with tool-using capabilities.

llm-tool-registry

Implements a tool registry to make tool calling easier when using LLMs

llm-tool-runtime

Tool calling runtime for text-only LLMs with LangChain support

llm-tool-store

Library for interacting with the LLM Tool Store API

llm-toolbox

llm-toolbox is now LMtoolbox

llm-toolbridge

A unified tool calling interface for LLM providers

llm-toolkit-schema

The foundational shared event schema for the LLM Developer Toolkit

llm-toolpy

ToolPy is a Python module made to facilitate the creation of tools using LLMs.

llm-tools-hub

Instant integration of external tools into LLM applications via automatic function calling.

llm-tools-quickjs

JavaScript execution as a tool for LLM

llm-tools-server

Reusable Flask server for LLM backends (Ollama, LM Studio) with tool calling

llm-tools-simpleeval

Make simple_eval available as an LLM tool

llm-topology

LLM access to Topology

llm-unified-orchestrator

Orchestrate LLM tasks. Define workflows, run and track progress.

llm-utility-pack
llm-utils

Utilities for our LLM projects (CWhy, ChatDBG, ...).

llm-wf

Moteur de workflow LLM basé sur des graphes d'agents

llm-workers

Core library for experimenting with LLMs - provides configuration, workers, tools, and LLM integrations

llm-works

LLM utilities and tools

llm-wrangler

LLM utilities for handling inputs and outputs.

llm00

An interface that allows you to use LLMs in an ultra-simple way

llm4chat

Create a Python package.

llm4gpt

Create a Python package.

llm7
llmFunctionDecorator

A simplified function decorator for OpenAI and LiteLLM API calls.

llmFunctionWrapper

A simplified function wrapper for OpenAI and LiteLLM API calls.

llm_batch_helper

A Python package that enables batch submission of prompts to LLM APIs, with simplified interface and built-in async capabilities handled implicitly.

llm_strategy

Directly Connecting Python to LLMs - Dataclasses & Interfaces <-> LLMs

More in Agent frameworks & infrastructure