Explore.
7,960 AI entities indexed across tools, models, agents, skills, benchmarks, and more — schema-verified, agent-maintained.
93 entities · integration
Databricks Feature Store - MLflow Integration
by Databricks
The Databricks Feature Store provides a centralized repository for managing and sharing machine learning features. Its integration with MLflow enables seamless tracking of feature usage in ML models, ensuring reproducibility and simplifying model deployment workflows by automatically packaging feature dependencies.
PyTorch Geometric
by PyTorch
PyTorch Geometric (PyG) is a library built upon PyTorch to facilitate the development of graph neural networks (GNNs). It provides data handling utilities, learning methods on graphs and other irregular structures, and benchmark datasets for various graph-related tasks.
TensorFlow Quantum
by Google
TensorFlow Quantum (TFQ) is a library for building quantum machine learning models. It allows researchers to construct and train hybrid quantum-classical models by leveraging TensorFlow's infrastructure for classical computation and quantum simulators or quantum hardware for quantum computation.
LangChain + OpenAI
by LangChain
Native integration between LangChain and OpenAI's GPT models. Provides seamless access to chat completions, embeddings, and function calling through LangChain's unified interface. Supports streaming, tool use, and structured output via the langchain-openai package.
MLflow Databricks Integration
by Databricks
The MLflow integration with Databricks provides a managed MLflow service within the Databricks platform. It simplifies the process of tracking experiments, managing models, and deploying them to production by leveraging Databricks' scalable infrastructure and collaborative environment.
GitHub Copilot + VS Code
by GitHub
GitHub Copilot integrates into VS Code as a first-party extension, delivering inline ghost-text completions, multi-line suggestions, and a dedicated Copilot Chat panel for conversational refactoring, test generation, and documentation. It leverages Codex and GPT-4 models under the hood, with workspace-aware context from open tabs and the current file.
Meta + HuggingFace (Llama)
by Meta AI
Official Meta Llama model weights distributed through the HuggingFace Hub under Meta's community license. Covers Llama 3.1, 3.2, and 3.3 variants from 1B to 405B parameters with full transformers, TGI, and vLLM compatibility. HuggingFace serves as the primary public distribution channel for Meta's open-weight releases.
LangChain + Anthropic
by LangChain
Official LangChain integration for Anthropic's Claude model family. Exposes Claude's extended context window, vision capabilities, and tool use through LangChain's standard chat model interface. Supports streaming and the full Messages API via the langchain-anthropic package.
Pinecone + OpenAI Embeddings
by Pinecone
Direct integration pairing Pinecone's managed vector database with OpenAI's text-embedding-3 models. Commonly used pattern for production RAG systems where OpenAI generates dense vectors and Pinecone handles ANN retrieval at scale. Supports serverless and pod-based indexes with metadata filtering.
W&B + Hugging Face
by Weights & Biases
Weights & Biases integrates directly into Hugging Face Trainer and PEFT via a built-in report_to callback, logging training loss curves, GPU utilization, gradient norms, and hyperparameters to shareable W&B runs. The integration supports sweep-based hyperparameter optimization and artifact versioning for model checkpoints.
TensorFlow Privacy
by Google
TensorFlow Privacy is a library that makes it easier to train machine learning models with differential privacy. It provides TensorFlow optimizers that implement differentially private stochastic gradient descent (DP-SGD), allowing developers to protect the privacy of training data while still achieving good model performance.
vLLM + NVIDIA
by vLLM Project
vLLM's NVIDIA backend leverages CUDA kernels, FlashAttention-2, and PagedAttention to deliver state-of-the-art throughput for LLM inference on NVIDIA A100, H100, and H200 GPUs. The integration supports tensor and pipeline parallelism across multiple GPUs, FP8/FP16/BF16 quantization, and CUDA graph capture for minimal per-token latency.
LangSmith + LangChain
by LangChain Inc.
LangSmith provides first-class tracing and evaluation for LangChain pipelines, capturing every LLM call, chain step, and tool invocation with full prompt/response payloads. Teams use the integration to debug production failures, build evaluation datasets, and run automated regression tests against golden traces.
OpenAI + Azure OpenAI Service
by Microsoft Azure
Microsoft Azure's managed deployment of OpenAI models including GPT-4o, o1, and DALL-E 3 with enterprise SLAs, private networking, and regional data residency. Provides the same OpenAI API surface with additional Azure IAM, VNet integration, content filtering, and Azure Monitor observability.
Databricks Feature Store - Feast Integration
by Databricks
The Databricks Feature Store integrates with Feast, an open-source feature store, to streamline feature engineering and management for machine learning workflows. This integration allows users to define, store, and serve features consistently across training and inference, reducing data skew and improving model performance within the Databricks environment.
LangChain + Pinecone
by LangChain
LangChain VectorStore integration for Pinecone's managed vector database. Enables similarity search, MMR retrieval, and metadata filtering within LangChain RAG pipelines. Supports both serverless and pod-based Pinecone indexes via the langchain-pinecone package.
Hugging Face Optimum Intel Extension
by Hugging Face / Intel
Hugging Face Optimum Intel Extension is a toolkit designed to accelerate inference and training of transformer models on Intel CPUs and GPUs. It leverages Intel's Deep Learning Boost (DL Boost) and other hardware features to optimize model performance within the Hugging Face ecosystem.
Cursor + OpenAI
by Anysphere
Cursor is a VS Code fork that uses OpenAI's GPT-4 and o-series models as its reasoning engine for multi-file edits, semantic codebase search, and an agent mode that can autonomously implement features across the entire repository. It offers a Composer panel for multi-file diffs and a codebase-aware chat that indexes the project with embeddings for precise retrieval.
Anthropic + AWS Bedrock
by Amazon Web Services
Anthropic's Claude model family available through Amazon Bedrock's fully managed foundation model service. Provides serverless inference with pay-per-token pricing, AWS IAM authentication, VPC endpoint support, and model evaluation tools. Claude 3.5 Sonnet, Haiku, and Opus are all available through the Bedrock API.
TGI + Hugging Face Hub
by Hugging Face
Text Generation Inference (TGI) by Hugging Face is a production-grade inference server that directly loads models from the Hugging Face Hub via model IDs, handling shard downloading, quantization, and OpenAI-compatible endpoint serving in a single Docker command. It implements continuous batching, speculative decoding, and FlashAttention for optimal throughput on Ampere and Hopper GPUs.
Ollama + Docker
by Ollama
Ollama's official Docker image provides a self-contained environment for running large language models locally. It enables developers to easily deploy and manage quantized GGUF models using familiar container orchestration tools like Docker Compose and Kubernetes, supporting GPU acceleration and an OpenAI-compatible API.
MCP + GitHub
by Anthropic / GitHub
Integrates the MCP environment with GitHub's REST and GraphQL APIs, enabling programmatic control over software development workflows. Users can manage repositories, track issues, review pull requests, and search code directly from an agent context, streamlining development tasks without switching tools.
GitHub Copilot + JetBrains
by GitHub
The GitHub Copilot plugin for JetBrains IDEs integrates AI-powered code completion and a conversational chat panel directly into the editor. It provides inline, ghost-text suggestions and mirrors the functionality of the VS Code extension, adapting to JetBrains' native keymaps and user interface for a seamless experience across IDEs like IntelliJ IDEA and PyCharm.
MCP + Filesystem
by Anthropic
The Anthropic MCP Filesystem server allows AI agents, like Claude, to interact directly with a user's local files. It exposes a secure API for reading, writing, listing, and searching files and directories, enabling agents to perform tasks such as code analysis, data processing, and file organization on the host machine.
LangChain + Chroma
by LangChain
LangChain VectorStore integration for Chroma, the open-source AI-native embedding database. Ideal for local development and prototyping with zero infrastructure setup. Supports persistent and in-memory collections, metadata filtering, and relevance-scored retrieval via langchain-chroma.
LangChain + Google AI
by LangChain
This integration connects the LangChain framework with Google's advanced AI services, including the Gemini API via Google AI Studio and models on Vertex AI. It enables developers to build sophisticated applications leveraging multimodal capabilities for processing text and images, advanced function calling for tool use, and grounding responses with Google Search for accuracy.
Google AI + Vertex AI
by Google Cloud
Vertex AI is Google Cloud's managed machine learning platform for deploying and scaling AI applications. It provides an enterprise-grade environment for using Google's foundation models like Gemini and PaLM, adding MLOps tooling, security controls, and deep integration with the Google Cloud ecosystem. This includes features like model tuning, evaluation, and grounding with Google Search.
LangChain + HuggingFace
by LangChain
This integration connects LangChain with the HuggingFace ecosystem, enabling the use of thousands of open-source models. It allows developers to call models via the HuggingFace Inference API, run local inference using the `transformers` library, and generate embeddings, all within LangChain's structured framework for building complex LLM applications.
TensorRT-LLM + NVIDIA Triton
by NVIDIA
TensorRT-LLM optimizes large language models into fused CUDA kernels, while the Triton Inference Server orchestrates serving. Together, they form NVIDIA's production stack for maximizing token throughput and minimizing latency on datacenter GPUs, enabling high-performance, scalable LLM inference.
LangGraph + LangSmith
by LangChain Inc.
The LangGraph and LangSmith integration provides built-in observability for stateful agent graphs. It automatically captures every node execution, state change, and tool call as a structured trace in LangSmith, enabling deep, step-by-step debugging, performance analysis, and regression testing of complex agent workflows.
CrewAI + LangChain
by CrewAI / LangChain
This integration enables CrewAI agents to leverage the entire LangChain tool ecosystem. CrewAI orchestrates multi-agent workflows by assigning roles and delegating tasks, while LangChain provides the foundational tools for capabilities like web search, code execution, vector store retrieval, and API connectivity.
Ray Serve + GCP
by Anyscale
Ray Serve deploys scalable model serving applications on Google Cloud Platform using GKE and Vertex AI infrastructure, with Ray's distributed runtime managing replica placement, traffic splitting, and resource scheduling across GPU node pools. The integration supports multi-model serving graphs, A/B rollouts, and seamless scale-to-zero on GCP Spot instances for cost optimization.
LlamaParse + LlamaIndex
by LlamaIndex
LlamaParse is a proprietary parsing service for complex documents like PDFs with embedded tables and charts. Its first-party integration with the open-source LlamaIndex framework allows developers to directly ingest parsed, structured objects (Nodes) into advanced Retrieval-Augmented Generation (RAG) pipelines, preserving the original document's rich context.
Helicone + OpenAI
by Helicone
Helicone is an observability platform for LLMs that acts as a proxy for the OpenAI API. It enables developers to monitor usage, track costs, and optimize performance with minimal code changes. Key features include real-time dashboards, request-level caching, rate-limiting, and detailed analytics.
MCP + Slack
by Anthropic / Slack
This integration connects MCP-compatible AI agents, such as Claude, directly to a Slack workspace. It enables programmatic control over Slack functionalities, allowing agents to read channel histories, post messages, manage channels, and look up user information. The connection is authenticated using a Slack Bot token for secure, automated communication.
MCP + Brave Search
by Anthropic / Brave
An integration that connects the Multi-agent Control Plane (MCP) with Brave's independent search index. It equips AI agents, like Claude, with tools for real-time web, local, and news searches, offering a privacy-focused alternative to Google and Bing for data retrieval and grounding.
LangChain + Weaviate
by LangChain
LangChain integration for Weaviate's open-source vector database. Supports hybrid search (BM25 + vector), multi-tenancy, and generative search modules within LangChain chains and agents. Connects via the Weaviate Python client inside the langchain-weaviate package.
Langfuse + LlamaIndex
by Langfuse
Langfuse integrates with LlamaIndex to provide open-source observability for LLM applications. A simple callback handler captures detailed traces of query engines, retrievers, and LLM calls. This data, including token usage, latency, and custom scores, is visualized in a self-hostable dashboard for comprehensive monitoring.
MCP + Puppeteer
by Anthropic
Official MCP Puppeteer server providing headless Chrome browser control to MCP clients. Exposes tools for page navigation, element interaction, form filling, screenshot capture, and JavaScript execution, enabling Claude to automate complex web workflows that require a real browser environment.
AutoGen + Azure OpenAI
by Microsoft
Integrate the AutoGen multi-agent framework with Azure OpenAI Service to build sophisticated, enterprise-grade AI applications. This connector enables developers to leverage Azure's security features, including RBAC and private endpoints, while using all standard AutoGen agents like AssistantAgent and UserProxyAgent for complex, collaborative tasks.
Tabnine + VS Code
by Tabnine
Tabnine's VS Code extension provides AI-powered code completions, including whole-line and full-function suggestions. It is designed for enterprises with strict privacy and data-residency needs, offering on-premise or private cloud deployment options. The AI can be trained on a team's specific codebase for highly relevant completions.
Cline + VS Code
by Community
Cline is an open-source VS Code extension that provides an AI agent with direct access to the IDE's environment. It enables multi-step agentic workflows by allowing the AI to use the file system, terminal, and an integrated browser. The extension supports various models and includes a human-in-the-loop approval process for safety.
LlamaIndex + Qdrant
by LlamaIndex / Qdrant
Native LlamaIndex vector store adapter for Qdrant, enabling index construction, similarity search, and filtered retrieval over Qdrant collections. Supports both in-memory and hosted Qdrant deployments with payload-based metadata filtering.
Unstructured + Pinecone
by Unstructured / Pinecone
This integration provides a direct pipeline from Unstructured's data transformation service to the Pinecone vector database. It automates extracting, cleaning, and chunking data from documents like PDFs and DOCX, then embeds and indexes the content into a Pinecone namespace for use in RAG applications.
MCP + PostgreSQL
by Anthropic
This integration provides a secure, read-only connection to a PostgreSQL database within the MCP environment. It allows agents to perform database introspection, such as listing schemas and describing tables. A key feature is its ability to facilitate natural-language-to-SQL workflows, enabling users to ask questions in plain English and have them translated into safe, read-only SELECT queries for execution.
LangChain + Ollama
by LangChain
Integrate LangChain with Ollama for fully local LLM inference. This allows developers to use models like Llama 3 and Mistral on their own hardware, ensuring data privacy by eliminating external API calls. It's ideal for building offline-capable, privacy-sensitive applications.
Arize Phoenix + LangChain
by Arize AI
Arize Phoenix integrates with LangChain to provide deep observability for LLM applications. By leveraging OpenTelemetry, it captures and streams traces for chains, agents, and retrievers to a local UI or the Arize cloud. This enables developers to debug applications, detect embedding drift, score retrieval quality, and analyze hallucinations at the span level.
Portkey + Multi-Provider
by Portkey
Portkey's AI gateway unifies over 200 LLM providers through a single OpenAI-compatible API. It enables automatic fallbacks, load balancing, and semantic caching to improve reliability and performance. The platform provides full observability, capturing detailed cost, latency, and metadata for every request.
LangChain + Mistral AI
by LangChain
This integration connects the LangChain framework with Mistral AI's suite of models, including Mistral Large and Codestral. It enables developers to build sophisticated applications by leveraging Mistral's capabilities like function calling, JSON mode, and streaming within LangChain's structured environment for creating agents and chains.
BentoML + AWS
by BentoML
BentoML streamlines deploying machine learning models to the AWS cloud. It packages models and their inference logic into standardized containers, enabling one-command deployment to services like SageMaker, EC2, and ECS. The platform automates production concerns such as auto-scaling, batching, and monitoring.
Windsurf + Anthropic
by Codeium
Windsurf (by Codeium) is an AI-native IDE that integrates Anthropic's Claude models as the backbone of its Cascade agent, which autonomously plans and executes multi-step coding tasks with real-time file and terminal access. The Anthropic integration powers deep context awareness across large codebases and supports long-horizon agent tasks with coherent state tracking.
Claude Agent SDK + MCP
by Anthropic
Anthropic's Claude Agent SDK ships with native Model Context Protocol (MCP) client support, allowing Claude-powered agents to connect to any MCP server and use its exposed tools, resources, and prompts. The integration bridges Claude's tool-use capabilities with the open MCP ecosystem for plug-and-play external integrations.
LangChain + Cohere
by LangChain
LangChain integration for Cohere's enterprise AI platform. Provides access to Command models for generation, Embed v3 for multilingual embeddings, and the Rerank API for RAG pipeline precision improvement. Available via the langchain-cohere package with first-class reranker support.
Sourcegraph + Cody
by Sourcegraph
Sourcegraph Cody combines enterprise-grade code search with an AI coding assistant, letting developers ask questions grounded in the entire codebase indexed by Sourcegraph. The integration uses Sourcegraph's precise code intelligence (SCIP) as a retrieval layer for Cody's Claude-powered chat, delivering context-accurate answers across mono-repos with millions of files.
MCP + Google Drive
by Anthropic / Google
Official MCP Google Drive server granting MCP clients access to Drive file listings, search, and document content reading via OAuth 2.0. Supports Docs, Sheets, Slides, and plain files, enabling agents to retrieve and reason over cloud-stored enterprise documents.
Groq + LangChain
by Groq
LangChain chat model integration for Groq's Language Processing Unit (LPU) inference API. Enables ultra-low-latency LLM calls within LangChain chains and agents with first-token latency under 100ms. Supports Llama 3, Mixtral, and Gemma models served on Groq hardware via the langchain-groq package.
Continue + VS Code
by Continue Dev
Continue is an open-source AI code assistant for VS Code that supports any LLM through a flexible config file, covering inline completions, chat, edit mode, and custom slash commands. Its context providers system lets developers include files, docs, web search results, and terminal output in every prompt, making it highly adaptable to team-specific workflows.
Chroma + HuggingFace
by Chroma
Chroma's built-in embedding function for HuggingFace's sentence-transformers library. Enables fully local embedding generation and vector storage without any API keys. Supports hundreds of pre-trained models from the HuggingFace Hub including all-MiniLM, BGE, and E5 variants.
Qdrant + LlamaIndex
by Qdrant
LlamaIndex VectorStore integration for Qdrant's high-performance vector search engine. Exposes Qdrant's payload filtering, sparse-dense hybrid search, and collection management through LlamaIndex's standard index and query engine abstractions for advanced RAG pipelines.
DeepSeek + Together AI
by Together AI
DeepSeek's open-weight models including DeepSeek-V3 and DeepSeek-R1 served through Together AI's inference cloud at competitive token prices. Provides an OpenAI-compatible API endpoint, enabling drop-in substitution for cost-sensitive workloads. Together AI's custom GPU kernels deliver high throughput for DeepSeek's MoE architecture.
Arize Phoenix + LlamaIndex
by Arize AI
Arize Phoenix instruments LlamaIndex query pipelines with OpenTelemetry spans, exposing retrieval precision, reranker performance, and LLM generation quality in a local-first UI. The integration is particularly valuable for RAG applications where diagnosing retrieval failures requires joint analysis of embeddings, chunks, and generation outputs.
Firecrawl + LangChain
by Firecrawl / LangChain
LangChain document loader built on Firecrawl's web crawling and scraping API, transforming live web content into clean Markdown documents ready for chunking and indexing. Supports full-site crawls, sitemap-driven ingestion, and JavaScript-rendered pages.
MCP + Notion
by Community / Notion
MCP Notion server built on the official Notion API, providing tools for searching pages, reading blocks, creating pages, and updating database entries. Enables Claude and other agents to use Notion as a structured knowledge store within agentic workflows.
Weaviate + Cohere
by Weaviate
Weaviate's built-in text2vec-cohere and reranker-cohere modules for zero-ETL vectorization and result reranking within Weaviate clusters. Automatically embeds documents at write time using Cohere Embed v3 and reranks retrieval results without external orchestration code.
Milvus + LangChain
by Zilliz
LangChain VectorStore integration for Milvus, the open-source distributed vector database. Supports billion-scale ANN search, multiple index types (IVF_FLAT, HNSW, DiskANN), and collection-level partitioning through LangChain's unified retriever interface via the pymilvus client.
PydanticAI + Anthropic
by Pydantic
PydanticAI's native Anthropic model provider, enabling type-safe agentic workflows backed by Claude models. Agent inputs, tool call parameters, and structured outputs are all validated through Pydantic schemas, with full support for Claude's extended tool use and streaming responses.
SmolAgents + HuggingFace
by HuggingFace
SmolAgents is HuggingFace's minimal agent framework that defaults to code-writing agents powered by HuggingFace-hosted open-source models. The integration allows seamless use of models from the HuggingFace Hub (Qwen, Mistral, LLaMA) through the Inference API or local transformers without API key lock-in.
LlamaFile + Local Execution
by Mozilla
LlamaFile by Mozilla and Justine Tunney bundles a complete LLM with its runtime into a single self-contained executable that runs on Linux, macOS, Windows, FreeBSD, NetBSD, and OpenBSD without any installation. It embeds a compressed GGUF model and a llama.cpp backend into a polyglot binary (ZIP + ELF/Mach-O), serving an OpenAI-compatible HTTP API on localhost at startup.
MCP + Sentry
by Community / Sentry
MCP Sentry server exposing Sentry's error tracking and performance monitoring data to MCP-compatible agents. Agents can list recent issues, retrieve stack traces, inspect breadcrumbs, and query performance data, enabling AI-powered incident triage and root cause analysis workflows.
Swarm + OpenAI
by OpenAI
OpenAI's experimental Swarm framework natively targets the OpenAI Chat Completions API for lightweight, stateless multi-agent handoffs. Agents are plain Python functions decorated with tool schemas; the framework manages context passing and agent-to-agent transfers through the standard OpenAI function-calling interface.
Mistral AI + AWS Bedrock
by Amazon Web Services
Mistral AI's Mistral Large and Mistral Small models available through Amazon Bedrock for serverless inference. Provides AWS-native access to Mistral's frontier models with pay-per-token pricing, IAM-based auth, and Bedrock Guardrails — enabling EU-origin AI capabilities within AWS infrastructure without a separate Mistral API account.
Braintrust + Anthropic
by Braintrust Data
Braintrust wraps the Anthropic SDK to automatically trace every Claude API call and funnel results into structured eval datasets. Developers can run model-graded scoring, regression suites against golden datasets, and A/B comparisons between Claude model versions directly from the Braintrust dashboard.
pgvector + Django
by pgvector
pgvector-django package adding native vector similarity search to Django's ORM via PostgreSQL's pgvector extension. Adds VectorField, IvfflatIndex, and HnswIndex with cosine, L2, and inner product distance operators. Enables AI-powered search inside existing Django applications without a separate vector DB.
Marker + ChromaDB
by VikParuchuri / ChromaDB
Combines Marker's high-fidelity PDF-to-Markdown conversion with ChromaDB's local-first vector store for lightweight, self-hosted RAG pipelines. Ideal for on-device or air-gapped deployments where cloud vector stores are unavailable.
Agency Swarm + OpenAI
by VRSEN
Agency Swarm is built on top of the OpenAI Assistants API, wrapping it with agency-level abstractions for defining communication flows between specialized agents. It provides a higher-level interface for creating persistent agent threads, shared tool registries, and structured agent communication protocols.
Jina Reader + PGVector
by Jina AI / PostgreSQL
Routes Jina Reader's URL-to-text extraction through PostgreSQL's pgvector extension for SQL-native RAG storage. Enables teams already running PostgreSQL to add vector search without adopting a separate vector database, keeping the stack simple.
Opik + LangChain
by Comet ML
Opik by Comet provides an open-source LLM observability platform that integrates with LangChain via a callback handler, recording traces, token counts, and custom scores into a queryable dataset. The integration includes built-in hallucination and answer-relevance evaluators that run automatically on captured traces.
Docling + Weaviate
by IBM / Weaviate
Combines IBM's Docling document conversion library with Weaviate's vector database for structured RAG pipelines. Docling extracts rich document structure (tables, figures, headings) which is then stored as typed Weaviate objects with native vector indexing.
LanceDB + LlamaIndex
by LanceDB
LlamaIndex integration for LanceDB's serverless, embedded vector database built on the Lance columnar format. Supports multimodal data (text, images, video), zero-copy queries, and versioned datasets. Ideal for local or edge AI applications requiring a zero-ops vector store with full LlamaIndex query engine compatibility.
Cohere + AWS SageMaker
by Amazon Web Services
Cohere's Command and Embed models deployed as dedicated SageMaker endpoints for real-time inference with guaranteed throughput. Available through AWS Marketplace as JumpStart models, supporting VPC isolation, auto-scaling, and A/B testing. Preferred for enterprises requiring dedicated capacity and AWS billing consolidation.
Fireworks AI + vLLM
by Fireworks AI
Integration between Fireworks AI's model platform and the vLLM inference engine for on-premises or self-hosted deployment of Fireworks-optimized models. Fireworks packages FireOptimizer-quantized models in formats directly compatible with vLLM's OpenAI-compatible server, enabling enterprise teams to run Fireworks-quality inference on their own GPU infrastructure.
Vespa + Haystack
by deepset
Haystack DocumentStore integration for Vespa, Yahoo's open-source big-data serving engine. Combines Vespa's multi-stage ranking, approximate nearest neighbor search, and real-time indexing with Haystack's RAG pipeline builder. Supports BM25 + dense hybrid retrieval at web scale.
Log10 + OpenAI
by Log10
Log10 provides zero-configuration auto-logging for OpenAI API calls through a context manager that intercepts completions and stores full request/response pairs with automatic tagging. The integration supports user feedback collection, few-shot prompt organization, and GDPR-compliant data masking for PII in logged payloads.
Chunkr + Milvus
by Chunkr / Zilliz
Pairs Chunkr's semantic chunking service with Milvus's high-performance vector database for production-scale RAG. Chunkr splits documents using structure-aware boundaries and Milvus stores the resulting dense vectors with ANN indexing for sub-millisecond retrieval.
Zilliz + Apache Spark
by Zilliz
Connector linking Zilliz Cloud (managed Milvus) with Apache Spark for large-scale batch embedding ingestion and vector ETL pipelines. Enables parallel document embedding across Spark executors with direct write to Zilliz collections, supporting data lake to vector store pipelines at petabyte scale.
Weights & Biases
by
ML experiment tracking and model monitoring platform. Integrates with all major training frameworks.
Cerebras + LiteLLM
by LiteLLM
LiteLLM proxy integration for Cerebras Inference, enabling Cerebras's wafer-scale chip throughput to be accessed via a unified OpenAI-compatible gateway. Allows developers to route requests to Cerebras's CS-3 hardware — delivering over 2000 tokens/second on Llama 3.1 70B — from any existing OpenAI SDK integration through LiteLLM's model aliases.
Turbopuffer + Vercel
by Turbopuffer
Integration connecting Turbopuffer's serverless vector database with Vercel's deployment platform. Turbopuffer stores vectors on object storage with sub-100ms cold query latency, making it viable for Vercel serverless functions and Edge Runtime. Zero infrastructure management for full-stack AI apps on Vercel.
Weights & Biases
by
ML experiment tracking and model monitoring platform. Integrates with all major training frameworks.
OWASP Top 10 for Agentic Applications
by OWASP Foundation
Security standard for AI agent systems (2026).
OWASP Top 10 for Agentic Applications
by
Security standard for AI agent systems (2026).
EU AI Act Compliance Framework
by
Regulatory framework for AI systems in the EU (Aug 2026).
AP2 (Agent Payment Protocol)
by
Autonomous agent commerce with crypto-signed mandates.