AI tools directory

A2A Python SDK

The A2A Python SDK implements the Agent2Agent protocol for discovering agents and exchanging tasks and messages.

Aegis

Aegis enforces policies for AI agents, providing guardrails and audit trails for their actions.

agent-safe-pipeline

Agent Safe Pipeline provides a security-focused framework for building AI agent workflows with policy checks and safety controls.

agentacct

The agentacct dashboard records coding-agent work steps, tools, file changes, tests, time, and token costs locally.

Agentic Security

Agentic Security scans AI agents and LLM applications for prompt injection, insecure tool use, and other vulnerabilities.

Agno (formerly Phidata)

Agno is an agent framework and runtime that orchestrates and runs multi-agent systems.

ai-gateway

AI Gateway routes AI model requests through a self-hosted gateway with observability and security controls.

AI-Infra-Guard

AI-Infra-Guard is an AI red teaming platform that scans AI infrastructure for CVE vulnerabilities, MCP/Agent skill risks, jailbreak weaknesses, and API relay security issues.

ai-memory

Solution for long term memory for agent coding CLIs and to facilitate handoff between different agent vendors

ai-safe2-framework

Ai-Safe2-Framework provides governance, risk, and compliance controls for securing agentic AI systems and non-human identities.

Ai2 Models

Ai2 develops the Olmo, Molmo, Tulu, and OlmoEarth families of open language and multimodal models.

Aider

Terminal-based AI pair programming tool that edits code in your local repository using LLMs like Claude, GPT-4o, and DeepSeek.

Alibaba Qwen Code

An open-source AI coding agent that lives in your terminal.

Alibaba Qwen Models

Qwen is Alibaba's language and multimodal model family, including coding, vision, image, speech, and reasoning models.

All Hands AI OpenHands

OpenHands is an open-source software-development agent that writes code, runs commands, and browses the web.

Amazon Kiro

Kiro is an agentic development environment that uses spec-driven workflows to plan, build, and maintain software.

Amazon Kiro Crew

Kiro Crew is an open-source workspace for coordinating AI coding agents.

Amazon Nova

Amazon Nova is Amazon's family of foundation models, including multimodal understanding, generation, and speech models.

Amazon Q Developer

Amazon Q Developer is an AI-powered coding assistant that helps developers build, debug, and optimize applications on AWS through conversational guidance and code suggestions.

Anthropic

Anthropic provides Claude AI models and APIs for building applications that generate, analyze, and automate work.

Anthropic Claude Code

Claude Code is Anthropic's terminal coding agent that plans, edits, and tests code in local repositories.

Anthropic Python SDK

The Anthropic Python SDK provides typed Python clients for Claude APIs, streaming responses, and tool use.

Anthropic TypeScript SDK

The Anthropic TypeScript SDK provides typed JavaScript and TypeScript clients for Claude APIs, streaming, and tool use.

Anyscale

Anyscale provides a managed Ray platform for distributed AI development, training, batch processing, and model serving.

Anyscale Ray

Ray is a distributed compute engine for AI workloads, bundling Ray Train, Ray Serve, Ray Data, and Ray Tune on a common runtime.

Anysphere Cursor

Cursor is an AI-powered code editor built on VS Code that uses large language models to assist with writing, debugging, and refactoring code.

Apple MLX

MLX is Apple's array framework for machine learning on Apple silicon, with a NumPy-like API, unified memory, and lazy evaluation.

Apple MLX LM

MLX LM generates text, quantizes models, and fine-tunes language models on Apple silicon using MLX.

Arcade

Arcade provides an actions runtime that lets AI agents securely call external services with authorization and governance controls.

ArcBox

ArcBox runs containers, microVM agent sandboxes, and full Linux or macOS virtual machines on a Mac from one Rust daemon and CLI.

Arize Phoenix

Arize Phoenix is an open-source platform that monitors LLM application traces and evaluates their outputs.

assistant-ui

Assistant-ui provides React components and state management for chat interfaces with streaming, tool calls, and attachments.

Augment Code

Your engineers have agents. Your organization doesn't. Cosmos is the platform that closes the gap.

AutoGPT

AutoGPT automates multi-step workflows with continuous AI agents that users can build and deploy.

AWS Amazon Bedrock

Amazon Bedrock provides managed foundation-model inference, customization, knowledge bases, and APIs for generative AI applications.

AWS Amazon Bedrock AgentCore

Amazon Bedrock AgentCore provides managed agent runtimes, memory, identity, gateways, observability, and evaluation services.

AWS Amazon SageMaker AI

Amazon SageMaker AI provides managed tools and infrastructure for building, training, fine-tuning, and deploying machine-learning models.

AWS Context Ontology Accelerator

Context Ontology Accelerator provides an open-source semantic layer that helps AI agents make consistent, explainable decisions using formal ontologies and knowledge graphs.

AWS Strands Agents

Strands Agents is AWS's open-source SDK for production agents in Python and TypeScript, with a model-driven loop, tool integration, and deployment targets.

Axolotl AI Axolotl

Axolotl fine-tunes open language models from YAML configuration, supporting full fine-tuning, LoRA, QLoRA, and multi-GPU training recipes.

Backslash

Backslash detects and prioritizes reachable vulnerabilities in cloud-native application code and dependencies.

Baseten

Baseten deploys and serves machine learning models as autoscaling inference endpoints, with a model API catalog and dedicated GPU deployments.

beelzebub

Beelzebub is a low-code deception runtime that runs AI-powered honeypots simulating systems for attackers.

BentoML

BentoML packages models and AI applications into inference APIs, job queues, and multi-model pipelines, and deploys them to Kubernetes or BentoCloud.

BerriAI LiteLLM

LiteLLM is an AI gateway that calls over 100 model APIs in OpenAI format with cost tracking, guardrails, load balancing, and logging.

Bhimraj Yadav receipt-ocr

Receipt-ocr runs Tesseract OCR and LLMs to extract structured data from receipt images through a CLI or API.

bitsandbytes

bitsandbytes provides low-bit quantization, matrix operations, and memory-efficient optimizers for training and inference.

Black Forest Labs FLUX

FLUX is Black Forest Labs' family of visual generation and editing models.

Block Goose

Goose is an open-source AI agent with a desktop app and CLI for coding and other automated tasks.

Braintrust

Braintrust provides evaluation, tracing, and improvement workflows for AI applications.

browser-use

The browser-use Python library automates web browsers for AI agents.

Browserbase Stagehand

Stagehand is an SDK for browser agents that mixes deterministic Playwright code with natural-language actions, with caching and previews for repeatability.

ByteDance Trae

Trae is an AI-powered IDE (Integrated Development Environment) that helps developers write, debug, and manage code using built-in AI assistance.

camel-ai

CAMEL-AI provides an open-source framework for building and studying communicative AI agents.

Cerebras Inference

Cerebras Inference provides hosted language-model inference through shared APIs and dedicated endpoints on Cerebras hardware.

Certiv

Complete visibility and control over every AI agent locally installed on the endpoint. Discover, understand, and safeguard every AI action across your organization.

Charm Crush

Terminal-based AI coding assistant that integrates LLMs, LSPs, and MCP servers into your development workflow.

Chroma

Search infrastructure for AI

City2Graph

City2Graph builds graph representations of geospatial relationships for graph neural network and spatial network analysis.

Claudoscope

Claudoscope is a native macOS menu-bar app that tracks Claude Code sessions, estimates costs, scans secrets, and surfaces analytics.

ClearML

ClearML provides experiment management, data versioning, pipeline orchestration, and GPU scheduling for machine learning workloads.

Cline

Autonomous coding agent as an SDK, IDE extension, or CLI assistant.

Cloudflare Agents

The Cloudflare Agents SDK builds stateful agents on Workers and Durable Objects, with persistent state, scheduling, WebSockets, and MCP server support.

Cognition Devin Desktop

Devin Desktop is an agent-native code editor for developing software locally, remotely, and in containerized environments.

Cohere Models

Cohere develops Command, Aya, Embed, Rerank, transcription, and document-understanding models.

Comet Opik

Opik traces, evaluates, and monitors LLM applications and agentic workflows, with datasets, an LLM-as-judge metric library, and production dashboards.

Comfy Org ComfyUI

ComfyUI is a node-graph interface, API, and backend for running diffusion models, chaining generation steps into reusable visual workflows.

Command Code

The first AI coding agent that learns your coding taste. Powered by taste-1, a meta neuro-symbolic model. A modern alternative to Claude Code, Cursor, OpenCode, and Copilot.

Composio

Composio connects AI agents to external tools and services with managed authentication and integrations.

compressed-tensors

compressed-tensors extends safetensors to store and load quantized, sparse, and otherwise compressed model weights.

Confident AI DeepEval

DeepEval is a unit-testing framework for LLM outputs, running metrics such as faithfulness, answer relevancy, and hallucination against test cases in CI.

Continue

Open-source coding agent available as a CLI, VS Code extension, and JetBrains plugin for AI-assisted development.

CopilotKit

CopilotKit connects AI agents to application interfaces with shared state, tools, and generative UI components.

CopilotKit OpenBot

OpenBot runs self-hosted AI coworkers that each get their own browser, files, and gated tools, with every action decided and recorded.

Cotool

Cotool is an AI security operations platform for investigating alerts, enriching threat intelligence, and automating response workflows.

CrewAI

Open-source Python framework for building multi-agent workflows using role-based AI agent Crews and event-driven Flows.

crust

Crust intercepts AI agent tool calls and blocks dangerous file, shell, and network actions before they execute.

CVE MCP Server

CVE MCP Server provides an MCP interface for looking up Common Vulnerabilities and Exposures (CVEs).

Daily Pipecat

Pipecat is a framework for realtime voice and multimodal agents, orchestrating speech-to-text, model inference, and text-to-speech as a streaming pipeline.

Databricks AI

Databricks AI provides managed model serving, agent development, evaluation, and retrieval over data in the Databricks platform.

Daytona

Daytona provisions isolated development sandboxes for AI coding agents through an API and SDK.

DeepInfra

DeepInfra provides hosted model inference APIs, private model deployments, and GPU infrastructure for language, image, video, and audio workloads.

DeepSeek Harness

DeepSeek Harness is an open-source agent harness for building coding agents with plugins.

DeepSeek Models

deepset Haystack

Haystack builds language-model applications with modular pipelines for retrieval-augmented generation and agents.

DeepSpeed

DeepSpeed is a deep learning optimization library for distributed training and inference, providing ZeRO sharding, pipeline parallelism, and mixed-precision execution.

DeepTeam

DeepTeam tests and red-teams LLM applications by simulating attacks and evaluating security vulnerabilities.

Diagram Design

Diagram Design is a Claude Code skill that generates editorial HTML and SVG diagrams and imports draw.io or Mermaid sources.

DMLC XGBoost

Distributed gradient boosting library implementing parallel tree boosting (GBDT/GBM) for scalable machine learning across major distributed environments.

Doberman-Core

Doberman-Core is a runtime security tool that blocks unsafe or unintended AI agent inputs, outputs, and tool calls before execution.

Docker MCP Gateway

Docker MCP Gateway runs MCP servers as isolated containers and routes client requests through a Docker CLI plugin.

Docker Sandboxes

Docker Sandboxes runs AI agents in isolated Docker environments with controlled filesystem, network, and tool access.

Docling

Docling parses PDFs, Office files, and HTML into structured documents with layout, tables, and reading order preserved for downstream RAG pipelines.

Docling Graph

Docling Graph converts documents into Pydantic objects and builds directed knowledge graphs with explicit semantic relationships using LLM or VLM backends.

dottxt Outlines

Outlines constrains language model generation to a grammar, regular expression, or JSON schema so the output is structurally valid by construction.

Earendil Works Pi

AI agent toolkit: unified LLM API, agent loop, TUI, coding agent CLI

Eigen Labs Darkbloom

Private Inference Network on Idle Macs

EleutherAI LM Evaluation Harness

The LM Evaluation Harness runs standardized few-shot benchmarks against language models and is the harness behind most published open-model eval numbers.

emisar

Emisar lets AI agents securely run infrastructure actions through MCP, with approval controls that protect production systems.

eunomia-bpf AgentSight

eBPF-based observability tool that profiles AI agent activity by correlating LLM traffic, process execution, file access, and network events without SDK integra

EverMind AI EverOS

EverOS provides a local-first memory layer that stores and evolves context for AI agents.

EXO

Clusters multiple devices to run large AI models locally, distributing inference across them with automatic discovery and RDMA support.

EXXETA exxperts

The exxperts framework runs local-first AI agents with approval-gated memory, model-provider integrations, MCP tools, and web search.

F5 CalypsoAI

CalypsoAI tests AI systems for security, safety, and compliance risks before deployment.

FailproofAI

Observability and enforcement for AI agent harnesses. Capture every run and runtime reliability with policy enforcement. 40 built-in policies, a local dashboard, no account required with a generous free cloud plan

fal

fal provides hosted generative-media model APIs and a serverless GPU platform for deploying custom inference workloads.

Feynman

Feynman is a personal AI research assistant that searches papers and the web, builds literature reviews, and produces cited research reports.

Fireworks AI

Fireworks AI provides hosted inference, model fine-tuning, and deployment APIs for open-weight models.

FlashAttention

FlashAttention implements fast, memory-efficient exact attention kernels used by most current transformer training and inference stacks.

FlashInfer

FlashInfer provides GPU kernels for language-model inference, including attention, sampling, and quantized operations.

FuzzyAI

FuzzyAI tests AI agents by fuzzing tool calls to find security vulnerabilities and unsafe behavior.

garak

Garak scans language models and applications for vulnerabilities including prompt injection and data leakage.

ggml whisper.cpp

whisper.cpp is a C/C++ port of OpenAI's Whisper speech recognition model, running transcription on CPU and GPU with no Python runtime.

Giskard

Giskard tests and evaluates LLM applications for performance, bias, and security issues.

GitHub Copilot CLI

GitHub Copilot CLI is a terminal-based coding agent that can plan, write, and test code in local repositories.

Google ADK

The Agent Development Kit is Google's code-first Python toolkit for building, evaluating, and deploying agents, with built-in tools, sessions, and deployment to Vertex AI.

Google DeepMind Optax

Optax provides composable gradient transformations and optimization algorithms for JAX.

Google Flax

Flax provides neural-network APIs for building and training models with JAX.

Google Gemini API

Gemini API provides Gemini models for generating text, images, and other multimodal content in applications.

Google Gemini Enterprise Agent Platform

Gemini Enterprise Agent Platform provides Google's managed model training, inference, evaluation, and agent deployment services, formerly offered under Vertex AI.

Google gemini-cli

An open-source command-line AI agent for coding, task automation, and tool use with Gemini models.

Google Gemma

Gemma is Google's family of downloadable language and multimodal models, including specialist variants.

Google Gen AI JavaScript SDK

The Google Gen AI JavaScript SDK provides TypeScript and JavaScript interfaces to Gemini APIs and live model sessions.

Google Gen AI Python SDK

The Google Gen AI Python SDK provides Python interfaces to Gemini model generation, streaming, tools, and embeddings.

Google JAX

Python library for accelerator-oriented array computation and program transformation, supporting automatic differentiation, JIT compilation, and scaling across

Google Keras

Multi-backend deep learning framework supporting JAX, TensorFlow, PyTorch, and OpenVINO for building and training models.

Google MaxText

MaxText provides JAX implementations and distributed-training workflows for language models on TPUs and GPUs.

Google TensorFlow

Open-source end-to-end machine learning platform with Python and C++ APIs for building and deploying ML models.

Google Tunix

Tunix provides JAX-based language-model post-training workflows, including supervised fine-tuning, preference optimization, and reinforcement learning.

GPT Researcher

GPT Researcher runs autonomous research workflows over web and local sources and produces cited reports.

gram

Gram provides MCP servers and AI-agent tools from APIs with access controls and observability.

Graphify

Graphify converts code, docs, SQL schemas, configs, and PDFs into a queryable knowledge graph with explainable relationships.

gridctl

Gridctl is a command-line tool that manages and routes Model Context Protocol servers, skills, and tools.

Groq

Groq provides cloud inference for running large language models with low latency.

GrowGraph OntoCast

OntoCast generates RDF knowledge graphs from documents by co-evolving domain ontologies and fact graphs with an agentic LLM pipeline.

Harmonic Security

Govern AI wherever your workforce runs it. Detect shadow AI, prevent data leaks, and coach employees inline — across web, desktop, and agentic workflows." name="description"/> <meta content=

Helicone

Routing and monitoring for reliable AI apps - the LLMOps platform behind the fastest-growing AI companies.

herdr

Herdr is a coding-agent runtime that manages persistent terminal sessions across local and SSH machines.

Heretic

Heretic removes safety alignment from transformer language models automatically, tuning directional-ablation parameters to minimise refusals while preserving the original model's behaviour.

HeyGen HyperFrames

HyperFrames is an open-source HTML-to-video renderer that runs in AI-agent workflows.

holmesgpt

SRE Agent - CNCF Sandbox Project

Hugging Face

Hugging Face is an open platform for building, sharing, and deploying machine-learning models, datasets, and applications.

Hugging Face Accelerate

Accelerate launches and runs PyTorch training across CPUs, GPUs, and TPUs with automatic mixed precision and configurable FSDP or DeepSpeed backends.

Hugging Face Candle

Candle is a minimalist machine learning framework for Rust with CPU, CUDA, and Metal backends, aimed at serverless and embedded inference.

Hugging Face Datasets

Datasets is a Python library for loading, streaming, and processing datasets for machine learning.

Hugging Face Diffusers

Diffusers provides pretrained diffusion pipelines, schedulers, and models for image, video, and audio generation in PyTorch.

Hugging Face LightEval

LightEval evaluates language models across inference backends and records detailed sample-level benchmark results.

Hugging Face Optimum

Optimum integrates model optimization and hardware acceleration tools with Hugging Face Transformers.

Hugging Face PEFT

PEFT implements parameter-efficient fine-tuning methods including LoRA, QLoRA, prefix tuning, and adapters for large pretrained models.

Hugging Face Sentence Transformers

Sentence Transformers trains and runs embedding, retrieval, and reranking models, and is the reference implementation behind most open embedding models.

Hugging Face smolagents

Smolagents builds AI agents that use code and tools.

Hugging Face Tokenizers

Tokenizers provides tokenization algorithms for training and running language models, with a Rust implementation and Python bindings.

Hugging Face Transformers

Python library providing pretrained transformer models for inference and training across text, vision, audio, video, and multimodal tasks.

Hugging Face TRL

TRL trains transformer language models with reinforcement learning, covering supervised fine-tuning, reward modeling, PPO, DPO, and GRPO.

IBM Granite

Granite is IBM's family of language, reasoning, vision, speech, time-series, and embedding models.

InfiniFlow RAGFlow

RAGFlow is a retrieval-augmented generation engine built on deep document understanding, combining chunk-level citations with agentic workflows.

Instructor

Instructor returns validated structured outputs from language models by defining response shapes as Pydantic models with automatic retries.

Intel AutoRound

AutoRound quantizes language and multimodal models through weight-rounding optimization and exports them for inference engines.

Intel OpenVINO

OpenVINO optimizes and deploys AI inference across Intel CPUs, GPUs, and NPUs, with model conversion, quantization, and a generative AI runtime.

InternLM LMDeploy

LMDeploy compresses, deploys, and serves language and vision-language models with its TurboMind engine, offering quantization and an OpenAI-compatible API.

iOfficeAI OfficeCLI

OfficeCLI is the first and best Office suite purpose-built for AI agents to read, edit, and automate Word, Excel, and PowerPoint files. Free, open-source, single binary, no Office installation required.

Jan AI Jan

Jan runs language models locally and provides a desktop chat interface.

Khoj

Khoj is a personal AI assistant that searches documents and the web, runs research tasks, and schedules automations.

Knostic

Knostic is an AI access control platform that manages and enforces need-to-know permissions for enterprise AI systems like Microsoft Copilot.

KoboldCpp

KoboldCpp runs GGUF language models locally as a single-file application with a built-in web UI and API.

KServe

KServe is a Kubernetes-native inference platform serving predictive and generative models behind a standard interface, with autoscaling and canary rollouts.

KTransformers

KTransformers provides kernels for language-model inference and fine-tuning with heterogeneous CPU and GPU execution.

Kyutai pocket-tts

pocket-tts synthesizes speech from text with a model small enough to run on a CPU.

LanceDB

LanceDB is an open-source database that stores multimodal data and supports vector search.

LangChain

LangChain is an open-source framework that orchestrates applications powered by language models.

LangChain Deep Agents

Deep Agents provides an agent harness with planning, filesystem tools, subagents, and context management.

LangChain LangGraph

Low-level orchestration framework for building, managing, and deploying long-running, stateful AI agents.

LangChain LangSmith

LangSmith provides tracing, evaluation, and deployment tools for LLM applications.

Langflow

Langflow is a visual builder for agents and RAG workflows that exports each flow as an API or Python object for deployment.

Langfuse

Langfuse provides tracing, evaluation, and monitoring for LLM applications.

LangGenius Dify

Dify is an LLM application platform for building agentic workflows and RAG pipelines through a visual workspace, deployable to cloud or self-hosted.

Letta (formerly MemGPT)

Letta manages persistent memory for stateful AI agents.

Letta Code

Letta Code is a terminal coding agent with persistent memory and identity across development sessions.

LibreChat

LibreChat provides a self-hosted chat interface with multiple model providers, agents, tools, and retrieval over uploaded files.

Lightning AI LitGPT

LitGPT provides language-model implementations and command-line workflows for pretraining, fine-tuning, evaluation, and deployment.

Lightning AI PyTorch Lightning

PyTorch Lightning organizes PyTorch training loops and scales pretraining and fine-tuning across one to thousands of GPUs without changing model code.

LiveKit Agents

LiveKit Agents builds realtime voice and video AI agents on LiveKit's WebRTC infrastructure, with turn detection, interruption handling, and telephony.

LLaMA-Factory

LLaMA-Factory unifies efficient fine-tuning for over 100 language and vision-language models through a command line and web interface.

llama.cpp

Llama CPP is a C/C++ library and set of command-line tools that runs large language models locally.

LlamaIndex

LlamaIndex is an open-source framework that connects language models to data sources.

LLM Compressor

LLM Compressor quantizes and compresses language models for deployment with vLLM and other inference engines.

LLM Guard

LLM Guard is an open-source security toolkit that detects, sanitizes, and protects LLM interactions against prompt injection, sensitive data leakage, and toxic language.

llmfit

Hundreds of models & providers. One command to find what runs on your hardware.

LM Studio

LM Studio is a desktop application that runs local LLMs and provides model discovery, downloads, chat, and a local API server.

LMCache

LMCache reuses and shares language-model key-value caches across requests and serving instances to reduce repeated prefill work.

LocalAI

LocalAI runs language models locally through self-hosted, OpenAI-compatible APIs.

Mamba

Mamba provides selective state-space model implementations and kernels for sequence modeling.

mantis

Mantis provides an open-source framework for testing application security with AI agents.

Mastra

Mastra is a TypeScript framework for AI applications and agents, providing workflows, agent memory, RAG, tool calling, and evals.

MCP Python SDK

The MCP Python SDK provides clients and servers for exposing tools, resources, and prompts through Model Context Protocol.

MCP TypeScript SDK

The MCP TypeScript SDK provides TypeScript clients, servers, and transports for Model Context Protocol.

Mem0

Mem0 is a memory layer for AI agents that extracts, stores, and retrieves durable facts across sessions through a hosted or self-hosted store.

Mend.io Mend AI

Traditional application security misses AI risk. Standalone AI security tools miss your code. Mend.io closes the gap with unified security for AppSec and AI.

Meta Faiss

Faiss is a library for efficient similarity search and clustering of dense vectors, and is the index engine underneath many vector databases.

Meta Llama

mex

Mex indexes codebases through a code graph and supplies a repo-local Markdown wiki to AI coding agents.

Microsoft Agent Framework

The Microsoft Agent Framework builds, orchestrates, and deploys agents and multi-agent workflows in Python and .NET, consolidating AutoGen and Semantic Kernel.

Microsoft AutoGen

Microsoft AutoGen is an open-source framework for building multi-agent AI applications that can act autonomously or alongside people.

Microsoft FLAML

FLAML automates model selection and hyperparameter tuning under computational resource constraints.

Microsoft Foundry

Microsoft Foundry provides managed AI models, agent development, evaluation, observability, and deployment services on Azure.

Microsoft LightGBM

Gradient boosting framework using tree-based learning algorithms for distributed, efficient training on large-scale data.

Microsoft MarkItDown

MarkItDown converts documents, spreadsheets, presentations, and other files into Markdown for language-model workflows.

Microsoft ONNX Runtime

ONNX Runtime runs machine-learning models across hardware and operating systems with an open-source inference engine.

Microsoft Phi

Phi is Microsoft's family of small language, reasoning, and multimodal models.

Microsoft Semantic Kernel

Semantic Kernel provides an open-source SDK for building AI agents and integrating language models into C#, Python, and Java applications.

Milvus

Milvus is an open-source vector database designed for storing, indexing, and searching high-dimensional vector embeddings at scale.

MiniMax Models

MiniMax develops language, vision, speech, music, image, and video generation models.

Mintplex Labs AnythingLLM

AnythingLLM is a local-first desktop and server application that runs agents, RAG workspaces, and MCP tools against local or hosted models.

Mistral AI

Mistral AI provides open-source and commercial language models for developers to build AI applications with flexible, efficient inference.

MLflow

MLflow tracks, evaluates, and monitors machine learning and LLM applications, covering experiment tracking, tracing, prompt management, and a model registry.

MLX Serve

Native Zig inference server for Apple Silicon. MLX + GGUF. OpenAI- and Anthropic-compatible APIs. No Python. Drop-in alternative to LM Studio and Ollama on macOS.

Modal

Bring your own code, and run CPU, GPU, and data-intensive compute at scale. The serverless platform for AI and data teams.

Model Context Protocol Servers

MCP Servers provides reference server implementations for connecting AI applications to external tools and data.

Model Context Protocol Specification

MCP Specification provides a standard for connecting AI applications to external tools and data through the Model Context Protocol.

ModelScope MS-SWIFT

MS-SWIFT provides training, fine-tuning, reinforcement learning, evaluation, and deployment workflows for language and multimodal models.

Moonshot AI Kimi

Kimi is Moonshot AI's family of language, coding, reasoning, and multimodal models.

Mozilla AI llamafile

llamafile packages a model and its runtime into a single executable file that runs on macOS, Linux, and Windows without installation.

nanobot

Nanobot is a self-hosted personal AI assistant framework with tools, memory, scheduled tasks, and integrations with chat applications.

Nebius AI Cloud

Nebius AI Cloud provides GPU compute, managed Kubernetes, distributed training infrastructure, and serverless AI workloads.

NeMo Guardrails

NeMo Guardrails provides an open-source toolkit for adding programmable guardrails to LLM-based applications.

Neo4j Agent Memory

Agent Memory is a graph-native memory system for AI agents that stores conversations, builds knowledge graphs, and tracks reasoning traces with Neo4j.

Neo4j Labs Neocarta

Neocarta builds a Neo4j semantic layer that gives AI agents structured access to schemas, business glossaries, metrics, and query history.

NeuML txtai

txtai is an embeddings database for semantic search, LLM orchestration, and language model workflows, with sparse, dense, and graph indexes in one store.

NoLabs nono

The nono runtime runs AI agents in isolated environments with policy-controlled filesystem, network, and credential access.

Nous Research Hermes

A terminal-native, self-improving AI agent for autonomous coding and tasks, with persistent memory, agent-created skills, and a multi-platform messaging gateway.

NVIDIA Dynamo

Dynamo is a datacenter-scale distributed inference serving framework with disaggregated prefill and decode, KV-cache-aware routing, and multi-engine support.

NVIDIA Megatron Bridge

Megatron Bridge connects Hugging Face model workflows with Megatron Core for distributed model training and checkpoint conversion.

NVIDIA Megatron-LM

Megatron-LM trains transformer models at scale with tensor, pipeline, sequence, and context parallelism, and ships Megatron Core as a reusable library.

NVIDIA Model Optimizer

NVIDIA Model Optimizer provides quantization, pruning, distillation, and other model-compression techniques for deployment.

NVIDIA NeMo Agent Toolkit

NeMo Agent Toolkit connects agent workflows to tools and model providers with profiling, evaluation, and observability.

NVIDIA NeMo Curator

NeMo Curator builds and scales data-curation pipelines for text, image, video, and audio training datasets.

NVIDIA NeMo RL

NeMo RL provides distributed reinforcement-learning and preference-optimization workflows for training language models.

NVIDIA NeMo Speech

NeMo Speech provides model training and inference tools for speech recognition, speech synthesis, audio processing, and speech language models.

NVIDIA NemoClaw

NemoClaw is an NVIDIA reference stack for running AI agents such as OpenClaw and Hermes inside OpenShell sandboxes.

NVIDIA Nemotron

Nemotron is NVIDIA's family of reasoning, vision, speech, and retrieval models, with training recipes and deployment guides.

NVIDIA Object-Oriented Agents (NOOA)

Object-Oriented Agents provides a Pythonic framework for building AI agents.

NVIDIA OpenShell

OpenShell runs autonomous AI agents in policy-controlled sandboxes with isolated filesystem, network, and credential access.

NVIDIA SkillSpector

SkillSpector scans AI agent skills for vulnerabilities and malicious patterns before installation.

NVIDIA TensorRT-LLM

TensorRT-LLM compiles and serves large language models on NVIDIA GPUs with quantization, in-flight batching, and speculative decoding.

NVIDIA Transformer Engine

Transformer Engine provides optimized kernels and low-precision training support for Transformer models on NVIDIA GPUs.

NVIDIA Triton Inference Server

Triton Inference Server is an open-source platform for deploying and serving AI models at scale across CPUs and GPUs.

okf-memory OKF Agent Memory

Git-native persistent memory for AI coding agents. Implements Google OKF v0.2 with sub-300µs in-memory BM25 search, embedded MCP server, and progressive disclosure. Slashes token bloat by 80% with zero external databases or dependencies. Built in pure Go.

Old Coder

Old Coder is an open-source skill for coding agents that runs code through a test and verification gauntlet, producing test plans and evidence reports.

Ollama

Local large language model runner that downloads, serves, and chats with models through a CLI and REST API.

OmniRoute

OmniRoute puts hundreds of model providers behind one endpoint, with quota-aware fallback and prompt compression.

oobabooga's Text Generation WebUI (textgen)

Oobabooga Textgen runs local language models in a desktop app with chat, vision, tool-calling, web search, and an OpenAI-compatible API.

Open WebUI

Open WebUI is a self-hosted chat interface for local and remote models, with RAG over uploaded documents, tool calling, and multi-user access control.

open-kritt

Open-kritt is a self-hosted platform that orchestrates AI agents to find, deduplicate, rank, and validate vulnerabilities in code repositories.

OpenAI

OpenAI provides AI models and APIs for building applications that generate and reason with text, images, and audio.

OpenAI Agents SDK

OpenAI Agents SDK is an open-source framework that provides tools, handoffs, guardrails, and tracing for agentic applications.

OpenAI Codex CLI

OpenAI Codex CLI runs an agent in the terminal that reads, changes, and tests code in local repositories.

OpenAI Python SDK

The OpenAI Python SDK provides typed Python clients for OpenAI model APIs, streaming, and asynchronous requests.

OpenAI TypeScript SDK

The OpenAI TypeScript SDK provides typed JavaScript and TypeScript clients for OpenAI APIs and streaming responses.

OpenAI Whisper

OpenAI speech recognition model for multilingual transcription, translation, and language identification from audio files.

OpenClaw

Personal AI assistant platform that connects messaging channels, tools, and model providers through a gateway.

OpenCompass

OpenCompass evaluates language and multimodal models across datasets, inference backends, and benchmark metrics.

OpenNMT CTranslate2

CTranslate2 runs Transformer models with a C++ inference engine, quantization, and CPU and GPU optimizations.

OpenRLHF

OpenRLHF trains language models with reinforcement learning and preference optimization using Ray, vLLM, and DeepSpeed.

OpenRouter

OpenRouter is a routing gateway that exposes hundreds of models from many providers behind one OpenAI-compatible API with failover and unified billing.

OpenWhispr

OpenWhispr is a desktop dictation and meeting-transcription app that supports local speech models and user-configured cloud providers.

OpenWorker

OpenWorker automates desktop tasks across files, tools, and integrations with governed, audited AI agent actions.

Parakeet ASR Server

Parakeet ASR Server transcribes audio on CPUs or GPUs using NVIDIA Parakeet ONNX models through a Whisper-compatible API.

PartMode

PartMode builds browser-based parametric CAD on OpenCascade WebAssembly with sketches, assemblies, drawings, and typed agent workflows.

Perplexity API

Perplexity API Platform - build with the Router, Agent, Search, and Embeddings APIs. Real-time, web-wide research and Q&A capabilities for your products.

pgvector

pgvector adds vector similarity search to PostgreSQL with exact and approximate nearest-neighbour indexes, keeping embeddings in the primary database.

PicoClaw

PicoClaw is a Go-based personal AI assistant that runs tools and automations on resource-constrained devices.

Pinecone

Pinecone is a managed vector database that stores and queries embeddings for AI applications.

Prefect FastMCP

FastMCP is a Python framework for building MCP servers and clients, handling transport, schema generation, authentication, and server composition.

Preferred Networks Optuna

Optuna provides automated hyperparameter optimization with programmable search spaces, sampling algorithms, and trial pruning.

Promptfoo

Promptfoo tests, evaluates, and red-teams LLM applications to detect vulnerabilities and compare model outputs.

PromptLayer

PromptLayer is a platform for managing, evaluating, and observing prompts and LLM applications.

PydanticAI

PydanticAI is a framework that builds type-safe Python agents with dependency injection, model integrations, tools, and structured outputs.

PyTorch

Tensors and Dynamic neural networks in Python with strong GPU acceleration

PyTorch ExecuTorch

ExecuTorch runs PyTorch models on mobile, embedded, and edge devices through an ahead-of-time export path and a lightweight C++ runtime.

PyTorch torchao

torchao provides quantization, sparsity, and low-precision training and inference tools for PyTorch.

PyTorch TorchTitan

TorchTitan provides PyTorch-native model implementations and recipes for large-scale distributed training of generative models.

PyTorch torchtune

torchtune is PyTorch's native post-training library for fine-tuning, distillation, and RL on language models with composable recipes.

Qdrant

Qdrant - High-performance, massive-scale Vector Database and Vector Search Engine for the next generation of AI. Also available in the cloud https://cloud.qdrant.io/

ralforion OrionBelt Ontology Builder

Browser-based ontology workbench for OWL ontologies and SKOS vocabularies. Streamlit + rdflib, no Java, no Protégé. OWL-RL reasoning, 22 SKOS validation checks, read-only SPARQL console, bulk operations, gist upper-ontology starters, merge-aware imports, interactive vis-network graph.

Rath Team OpenRath

OpenRath provides an open-source platform for building AI agents that analyze data and create visualizations.

Rebuff

Rebuff detects and protects LLM applications against prompt injection attacks.

Red Hat ripwire

ripwire gives coding agents a ranked, deterministic map of a repository — call graph, blast radius, and tests to run — from a CLI or over MCP.

Replicate

Replicate is a platform for running and sharing machine learning models through a simple API without managing infrastructure.

Replit Agent

Replit Agent is an AI coding agent that builds, tests, and deploys applications in Replit.

RunPod

Explore our guides and examples to deploy your AI/ML application on Runpod. Review setup and usage guidance in the Runpod documentation.

Safetensors

Safetensors provides a tensor serialization format and libraries for storing and loading model weights without pickle execution.

SambaNova SambaCloud

SambaCloud provides managed model inference with OpenAI-compatible APIs on SambaNova hardware.

SAND Security

Security for the modern AI infrastructure. Govern, detect, and respond to threats across agents, untrusted code, and the AI cloud.

scikit-learn

Python machine learning library built on SciPy providing classification, regression, clustering, and model selection tools.

SecureAgentics Adrian

Adrian monitors AI agent actions and reasoning traces to detect and block malicious or out-of-remit behavior.

Semantica AGI Semantica

Semantica provides graph-native context infrastructure for building accountable AI systems.

SGLang

SGLang runs large language models through an inference server with efficient structured generation.

shell-gpt

ShellGPT is a command-line assistant that turns natural-language prompts into shell commands and code snippets.

sirmalloc ccstatusline

ccstatusline is a customizable statusline for Claude Code CLI with powerline support, themes, and widget-based configuration.

SkyPilot

SkyPilot runs AI training and inference jobs across clouds and Kubernetes clusters, picking cheaper or available capacity and handling spot preemption.

Skyvern

Skyvern automates browser-based workflows with vision models and agents, completing form-driven tasks on sites it has not seen before.

Snowflake Cortex AI

Snowflake Cortex AI provides managed language-model functions, search, and agents that work with data governed in Snowflake.

Snyk Agent Scan

Snyk Agent Scan is a command-line tool that scans AI agent skills and configurations for security risks.

Sourcegraph Amp

Amp is Sourcegraph's agentic coding tool for the terminal and editor, running multi-step edits with subagents and shared team threads.

SST OpenCode

OpenCode is an open-source AI coding agent for terminal-based development tasks.

Stability AI Models

Stability AI develops the Stable Diffusion, Stable Audio, Stable Video, and 3D model families.

StackBlitz bolt.new

Bolt.new is an AI-powered web development platform that builds websites and applications from prompts.

Stanford NLP DSPy

DSPy builds language-model pipelines and optimizes their prompts and weights.

stride-gpt

Stride-GPT generates threat models and attack trees with large language models and the STRIDE framework.

Superagent

Superagent is an open-source SDK that blocks prompt injections, redacts sensitive data, and scans repositories for threats targeting AI agents.

Superlog Labs Superlog

Superlog monitors AI agent telemetry with OpenTelemetry and groups signals into incidents for observability.

Superset

Superset is an agentic IDE to orchestrate 100+ coding agents in parallel. Run any agent with your own subscription.

SWE-agent

SWE-agent is an autonomous LM-powered agent for resolving GitHub issues and software-engineering tasks.

Tencent WeMM-Embedding

WeMM-Embedding is a family of multimodal embedding models for retrieval across text and images.

The Open Engine Zeroshot

Zeroshot orchestrates executor-verifier agent loops that implement and validate software changes while logging workflow state to SQLite.

ThHanke Ontosphere

Browser-based RDF/OWL knowledge graph editor with client-side OWL 2 DL reasoning, SHACL validation, reasoner-verified repair, and a 43-tool MCP server for AI ag

Thinking Machines Lab Tinker

Tinker provides a managed API for training and sampling language models while users control their training algorithms and data.

TileLang

TileLang is a Python-based domain-specific language and compiler for writing optimized GPU and CPU kernels.

timgordontg engrim

The Universal Cross-Model Episodic Memory Standard. Local-first, project-scoped SQLite memory engine for Google Antigravity, Claude Code, Cursor, Windsurf, and Codex. Zero cloud lock-in.

tiny corp tinygrad

tinygrad is a minimal deep learning framework built around lazy tensor operations and a small kernel-fusing compiler, targeting several accelerators.

Together AI

Run, train, and serve open-source AI models on Together AI.

ToolHive

ToolHive runs MCP servers in isolated containers and enforces identity and access policies for local and Kubernetes deployments.

Traceloop OpenLLMetry

OpenLLMetry instruments LLM and vector-database calls as OpenTelemetry spans so traces land in existing observability backends rather than a new one.

Trailhq Graft

Graft builds a code knowledge graph that gives coding agents compact, repository-specific context through a CLI and MCP server.

Triton

Triton is a Python-embedded language and compiler for writing GPU kernels, and is the backend most PyTorch and inference-engine kernels are generated through.

turbovec

Turbovec builds compressed approximate nearest-neighbor indexes in Rust with Python bindings using Google's TurboQuant algorithm.

typedef.ai fenic

Fenic runs DataFrame queries with AI operators for extracting, classifying, summarizing, embedding, and joining data.

Uber ADR

ADR discovers, monitors, benchmarks, and detects threats in AI agents such as Cursor, Claude Code, and Codex.

UK AI Security Institute Inspect AI

Inspect is the UK AI Security Institute's framework for large language model evaluations, covering datasets, solvers, scorers, and agentic eval tasks.

Unsloth AI Unsloth

Unsloth runs and fine-tunes language and diffusion models locally through a Python library and interface.

Unstructured

Unstructured is an open-source ETL library that converts PDFs, images, and office documents into chunked, structured elements for LLM ingestion.

Vane

Vane is a self-hosted AI answer engine that searches the web and composes responses with source citations.

vArmor

The vArmor Kubernetes runtime security platform provides AppArmor, Seccomp, and eBPF policy enforcement for workloads.

Vault-LD

Vault-LD stores knowledge graphs as Markdown with YAML front matter and maps them to linked data.

Vercel AI SDK

The Vercel AI SDK provides TypeScript APIs for model generation, structured output, tool use, and streaming application interfaces.

Vercel v0

The v0 generative UI platform builds web applications from natural-language prompts.

verl

verl is a reinforcement learning post-training framework for language models, implementing the HybridFlow programming model over Megatron and FSDP backends.

vLLM

The vLLM inference engine runs large language models with high-throughput, memory-efficient GPU execution.

vMLX

The vMLX server runs compressed MLX models on Apple Silicon with disk caching, paged memory, continuous batching, and hybrid SSM scheduling.

Volcengine OpenViking

OpenViking is a context database that manages agent memory, knowledge retrieval, and skills.

Warp

Warp is an AI terminal that runs commands and helps developers build software.

Weaviate

Weaviate is an open-source vector database that stores both objects and vectors, allowing for the combination of vector search with structured filtering with the fault tolerance and scalability of a cloud-native database​.

Weights & Biases

Weights & Biases tracks experiments, sweeps, artifacts, and model versions for training runs, and hosts the registry teams promote models through.

Weights & Biases Weave

Weights & Biases Weave traces, evaluates, and monitors LLM applications for developers.

Windsurf

xAI Grok API

Official SpaceXAI API documentation. Build with Grok: chat, image and video generation, voice, tool calling, and structured outputs. Quickstarts, guides, and API reference.

xalgorix

The xalgorix platform runs AI pentesting agents for reconnaissance, vulnerability detection, and exploitation workflows.

XGrammar

XGrammar constrains language-model output to grammars and schemas with optimized token-mask generation.

Z.ai GLM

GLM is Z.ai's family of language, reasoning, vision, image-generation, and speech models.

Zed

Zed is an open-source code editor that runs collaborative development workflows.

ZenML

ZenML is an MLOps and LLMOps framework for building portable pipelines that run the same code across local, cloud, and orchestrator backends.

Zep Graphiti

Graphiti builds temporally-aware knowledge graphs from conversations and documents so agents can query facts that change over time.

ZeroClaw

ZeroClaw runs personal AI assistants with configurable model providers, tools, memory, and messaging channels.

zvec-grep

zvec-grep searches a local workspace with combined keyword and semantic matching, from a CLI or over MCP.