Integrations
177 integrations. Zero config.
Auto-instrument every major AI framework, LLM provider, vector database, and tool in your stack with a single line of code.
141Auto-Instrumentors
23+Auto-Detected Providers
8Guardrail Frameworks
17Embedding Providers
8Rerankers
Showing 177 of 177 integrations
LLM Providers
🟢
OpenAI
GPT-4.1, o3-pro, o4-mini
Auto-instrumentor
🧠
Anthropic
Claude Opus 4.6, Sonnet 4.6
Auto-instrumentor
💎
Google Gemini
Gemini 3 Pro, Flash, Deep Think
Auto-instrumentor
☁️
AWS Bedrock
Converse API, InvokeModel
Auto-instrumentor
🔵
Azure OpenAI
Azure-hosted OpenAI models
Provider Detection
🌪️
Mistral AI
Mistral Large 3, Devstral 2
Auto-instrumentor
🔍
Cohere
Command A, Reasoning, Vision
Auto-instrumentor
⚡
Groq
Llama 4, Mixtral on LPU
Auto-instrumentor
🦙
Meta (Llama)
Llama 4 Scout, Maverick
Auto-instrumentor
🔬
AI21 Labs
Jamba models
Auto-instrumentor
🌐
DashScope / Alibaba Qwen
Qwen3 series models
Auto-instrumentor
💠
IBM WatsonX
Granite models
Auto-instrumentor
📊
Vertex AI
Google Cloud AI Platform
Auto-instrumentor
🔗
LiteLLM
100+ providers via unified API
Auto-instrumentor
🦙
Ollama
Local model serving
Auto-instrumentor
🤝
Together AI
200+ open-source models
Auto-instrumentor
Agentic Frameworks
🕸️
LangGraph
Stateful multi-actor agents
Callback Handler
🔗
LangChain
LLM application framework
Callback Handler
🦙
LlamaIndex
Data-connected LLM apps
Callback Handler
👥
CrewAI
Multi-agent orchestration
Auto-instrumentor
🤖
AutoGen
Microsoft Agent Framework
Auto-instrumentor
🧩
Semantic Kernel
Microsoft AI SDK
Auto-instrumentor
🟢
OpenAI Agents SDK
Official OpenAI agent framework
Auto-instrumentor
🟣
Claude Agent SDK
Anthropic agent framework
Auto-instrumentor
🧬
AWS Strands Agents
Model-agnostic agents
Auto-instrumentor
💎
Google ADK
Google Agent Development Kit
Auto-instrumentor
☁️
Amazon Bedrock Agents
Managed agent service
Auto-instrumentor
🧑💻
Agno
Formerly Phidata
Auto-instrumentor
✅
PydanticAI
Type-safe AI agents
Auto-instrumentor
🤗
smolagents
HuggingFace agent library
Auto-instrumentor
📐
DSPy
Programmatic prompt optimization
Auto-instrumentor
🌾
Haystack
NLP pipeline framework
Auto-instrumentor
🐫
CAMEL
Communicative agent framework
Auto-instrumentor
🧱
Composio
Tool integration for agents
Auto-instrumentor
🎓
Instructor
Structured LLM outputs
Auto-instrumentor
📝
Letta
Formerly MemGPT
Auto-instrumentor
🔄
ControlFlow
Structured AI workflows
Auto-instrumentor
🦙
Llama Stack
Meta agent framework
Auto-instrumentor
🎤
LiveKit Agents
Real-time voice AI agents
Auto-instrumentor
🤖
Langroid
Multi-agent task delegation
Auto-instrumentor
🦸
SuperAGI
Autonomous agent platform
Auto-instrumentor
🐝
Agency Swarm
Multi-agent orchestration
Auto-instrumentor
🔧
Mastra
TypeScript AI framework
Auto-instrumentor
📋
Julep
Stateful AI agent platform
Auto-instrumentor
🐱
Pipecat
Voice agent framework
Auto-instrumentor
Vector Databases
🌲
Pinecone
Managed vector database
Auto-instrumentor
🌐
Weaviate
Open-source vector search
Auto-instrumentor
📍
Qdrant
High-performance vector search
Auto-instrumentor
🐳
Milvus
Cloud-native vector DB
Auto-instrumentor
🌈
Chroma
AI-native embedding database
Auto-instrumentor
🏹
LanceDB
Serverless vector database
Auto-instrumentor
🔎
OpenSearch
Distributed search & analytics
Auto-instrumentor
🔍
Marqo
Tensor search engine
Auto-instrumentor
💨
Turbopuffer
Fast serverless vector DB
Auto-instrumentor
🐝
Vespa
Search & recommendation engine
Auto-instrumentor
🐘
PostgreSQL / pgvector
Postgres vector extension
Auto-instrumentor
🔴
Redis Vector Search
In-memory vector search
Auto-instrumentor
🍃
MongoDB Atlas Vector
Atlas vector search
Auto-instrumentor
🔍
Elasticsearch
Distributed search engine
Auto-instrumentor
📈
Neo4j
Graph database with vectors
Auto-instrumentor
📊
FAISS
Facebook AI Similarity Search
Auto-instrumentor
🦆
DuckDB Vector
In-process analytical database
Auto-instrumentor
👁️
Cassandra Vector
Distributed vector search
Auto-instrumentor
🌲
Annoy
Spotify ANN library
Auto-instrumentor
📊
hnswlib
HNSW vector search
Auto-instrumentor
🔍
USearch
Unum vector search engine
Auto-instrumentor
⚡
ScaNN
Google scalable nearest neighbors
Auto-instrumentor
🐳
Zilliz Cloud
Managed Milvus
Auto-instrumentor
⚡
Supabase
pgvector integration
Auto-instrumentor
💾
SingleStore
Vector functions
Auto-instrumentor
Graph Databases
🔴
FalkorDB
Redis-based graph database
Auto-instrumentor
🥑
ArangoDB
Multi-model graph database
Auto-instrumentor
🧠
Memgraph
In-memory graph database
Auto-instrumentor
🔱
Neptune
AWS managed graph database
Auto-instrumentor
📊
Microsoft GraphRAG
Knowledge graph-enhanced RAG
Auto-instrumentor
💡
LightRAG
Lightweight graph RAG
Auto-instrumentor
RAG Platforms
🛤️
Pathway
Real-time streaming RAG
Auto-instrumentor
🌊
RAGFlow
Deep document understanding
Auto-instrumentor
🚀
R2R (SciPhi)
Production REST API for RAG
Auto-instrumentor
🔍
Vectara
Managed RAG-as-a-Service
Auto-instrumentor
Embeddings
🟢
OpenAI Embeddings
text-embedding-3 models
Built-in
🔍
Cohere Embed
embed-v4 multilingual
Built-in
🚢
Voyage AI
Code & text embeddings
Auto-instrumentor
💠
Jina AI
Multimodal embeddings
Auto-instrumentor
💎
Google Gemini Embeddings
Gemini embedding models
Built-in
☁️
AWS Bedrock Titan
Titan embedding models
Built-in
🔵
Azure OpenAI Embeddings
Azure-hosted embeddings
Provider Detection
🌪️
Mistral Embeddings
Mistral embed models
Built-in
🗺️
Nomic
Open embedding models
Auto-instrumentor
🍞
MixedBread
Specialized embeddings
Auto-instrumentor
💬
sentence-transformers
SBERT framework
Auto-instrumentor
🏆
BGE (BAAI)
BAAI general embeddings
Auto-instrumentor
⚡
FastEmbed
Qdrant ONNX embeddings
Auto-instrumentor
🦙
Ollama Embeddings
Local embedding models
Built-in
📝
E5 (Microsoft)
Multilingual embeddings
Auto-instrumentor
🎓
Instructor Embeddings
Task-specific embeddings
Auto-instrumentor
🤗
TEI
HF Text Embeddings Inference
Auto-instrumentor
Rerankers
🔍
Cohere Rerank
Rerank 3.5 model
Built-in
💠
Jina Reranker
Cross-encoder reranking
Auto-instrumentor
🌲
Pinecone Rerank
Integrated reranking
Auto-instrumentor
🚢
Voyage Reranker
Voyage rerank models
Auto-instrumentor
⚡
FlashRank
Lightweight reranking
Auto-instrumentor
🏆
BGE Reranker
BAAI reranker models
Auto-instrumentor
🔄
CrossEncoder
sentence-transformers reranking
Auto-instrumentor
📑
ColBERT
Late interaction reranking
Auto-instrumentor
Guardrails
🛡️
Guardrails AI
Input/output validation
Auto-instrumentor
💚
NVIDIA NeMo Guardrails
Programmable guardrails
Auto-instrumentor
🔒
LLM Guard
Open-source LLM security
Auto-instrumentor
🚨
Lakera Guard
Prompt injection protection
Auto-instrumentor
🔵
Azure AI Content Safety
Content moderation API
Auto-instrumentor
🟢
OpenAI Moderation API
Content moderation
Auto-instrumentor
🕵️
Presidio
Microsoft PII detection
Auto-instrumentor
🛡️
PolyGuard
Multi-layer guardrails
Auto-instrumentor
Eval & Testing
🧪
DeepEval
LLM evaluation framework
Auto-instrumentor
📏
RAGAS
RAG evaluation toolkit
Auto-instrumentor
🧠
Braintrust
AI eval & observability
Auto-instrumentor
🔎
TruLens
LLM evaluation & tracking
Auto-instrumentor
🛡️
Giskard
ML testing & compliance
Auto-instrumentor
🔬
Inspect AI
UK AISI evaluation framework
Auto-instrumentor
🧪
promptfoo
CLI prompt testing & red-teaming
Auto-instrumentor
🔥
Arize Phoenix
Open-source observability + eval
Auto-instrumentor
☄️
Opik (Comet)
Tracing + evaluation
Auto-instrumentor
🔗
LangSmith
LangChain tracing + eval
Auto-instrumentor
📊
Langfuse
Open-source tracing + eval
Auto-instrumentor
Memory
🧠
Mem0
Long-term AI memory
Auto-instrumentor
📝
Letta (MemGPT)
Stateful agent memory
Auto-instrumentor
⌚
Zep
Conversation memory service
Auto-instrumentor
Model Serving
🚀
vLLM
High-throughput LLM serving
Auto-instrumentor
💡
SGLang
Structured generation runtime
Auto-instrumentor
🦙
Ollama (Self-Hosted)
Local model runtime
Auto-instrumentor
🦙
llama.cpp
C++ LLM inference
Auto-instrumentor
🤗
HuggingFace Transformers
Model library & pipeline
Auto-instrumentor
🤗
HuggingFace TGI
Text Generation Inference
Auto-instrumentor
🦎
ExLlamaV2
Quantized LLM inference
Auto-instrumentor
💠
TensorRT-LLM
NVIDIA max-performance inference
Auto-instrumentor
🔱
Triton Inference Server
NVIDIA multi-framework serving
Auto-instrumentor
💻
LM Studio
Desktop local inference
Auto-instrumentor
🏠
LocalAI
OpenAI-compatible local
Auto-instrumentor
📄
llamafile
Single-file LLM distribution
Auto-instrumentor
🍱
BentoML
Build, ship, scale AI apps
Auto-instrumentor
Multi-Modal
🎙️
Deepgram
Nova-3 STT & TTS
Auto-instrumentor
🎧
ElevenLabs
Voice AI, TTS & Scribe STT
Auto-instrumentor
🎤
AssemblyAI
Universal-2 transcription
Auto-instrumentor
💬
Whisper.cpp
Local speech recognition
Auto-instrumentor
▶️
PlayHT
800+ voice generation
Auto-instrumentor
📞
Vapi
Voice AI agent platform
Auto-instrumentor
🔊
Retell AI
Voice agent builder
Auto-instrumentor
🎨
Stable Diffusion
Open-source image generation
Auto-instrumentor
✨
Flux
Black Forest Labs image gen
Auto-instrumentor
🚀
Fal AI
Fast image generation API
Auto-instrumentor
Auto-Detected Providers
🌊
DeepSeek
V3.2, R1 reasoning models
Provider Detection
✨
xAI (Grok)
Grok 3, Grok 4.20
Provider Detection
🛤️
OpenRouter
300+ models via unified API
Provider Detection
🎆
Fireworks AI
Fast inference platform
Provider Detection
🔄
Replicate
Run open-source models
Provider Detection
🔮
Perplexity AI
Search-augmented LLMs
Provider Detection
💨
Cerebras
Wafer-scale inference
Provider Detection
🎯
SambaNova
Purpose-built AI hardware
Provider Detection
🚀
DeepInfra
Serverless GPU inference
Provider Detection
🌟
Novita AI
Model API platform
Provider Detection
🌌
Nebius AI
Cloud AI infrastructure
Provider Detection
📦
Baseten
Model serving infrastructure
Provider Detection
🟠
Cloudflare Workers AI
Edge AI inference
Provider Detection
⚛️
Lepton AI
AI application platform
Provider Detection
▶️
Modal
Serverless cloud compute
Provider Detection
🏃
RunPod
GPU cloud for AI
Provider Detection
❄️
Snowflake (Arctic)
Arctic enterprise models
Provider Detection
📎
Databricks (DBRX)
DBRX open models
Provider Detection
✍️
Writer (Palmyra)
Palmyra enterprise models
Provider Detection
🤝
FriendliAI
Optimized inference
Provider Detection
💠
SiliconFlow
Open-source model hosting
Provider Detection
Tools & Protocols
🔌
MCP
Model Context Protocol
Auto-instrumentor
🔥
Firecrawl
Web scraping for LLMs
Auto-instrumentor
🔧
OpenAI Function Calling
Structured tool invocation
Built-in
🛠️
Anthropic Tool Use
Claude tool calling
Built-in
📡
A2A Protocol
Google Agent-to-Agent protocol
Auto-instrumentor
🖥️
Computer Use
Anthropic browser automation
Built-in
How It Works
Three integration patterns
One line of code enables all of them. Waxell picks the right strategy for each library.
Auto-instrumentor
Monkey-patches the SDK at import time. Every call to the provider is automatically traced with zero code changes beyond the initial instrument_all() call.
141 integrations
Provider Detection
Automatically detects OpenAI-compatible API providers by inspecting the base_url. No additional SDK needed -- if it speaks the OpenAI protocol, Waxell sees it.
23+ providers
Callback Handler
Plugs into framework-native callback systems (like LangChain callbacks or LlamaIndex event handlers) for deep, step-level tracing of agent workflows.
3 frameworks
Get started in 30 seconds
One line instruments your entire stack. Every LLM call, every vector query, every agent step -- traced and governed automatically.
Python
from waxell_observe import instrument_all instrument_all() # That's it. 177 integrations activated.