Welcome
Quickstart
Serverless
AWS Lambda

Overview

Introduction

LLM Observability in Middleware provides comprehensive monitoring and troubleshooting capabilities for your LLM-powered applications. It offers deep insights through metrics, dashboards, and traces, enabling you to optimize performance and quickly resolve issues.

Key Features

  1. Traces: Get detailed, end-to-end tracing of your LLM requests and workflows.
LLM Main Page

  1. Metrics: Capture essential LLM-specific metrics:

  2. Dashboards: Utilize pre-built dashboards for quick insights into your LLM application's performance.

LLM Main Page

Benefits

  • Enhanced Visibility: Gain deep insights into your LLM application's performance and behavior.

  • Quick Troubleshooting: Identify and resolve issues faster with comprehensive tracing and metrics.

  • Performance Optimization: Use collected data to optimize your LLM applications for better efficiency and cost-effectiveness.

  • Seamless Integration: Easy integration with popular LLM providers and frameworks.

LLM Sidebar

Supported SDKs

Middleware supports two OpenTelemetry-compatible SDKs for LLM Observability:

Traceloop

Traceloop

OpenLIT

OpenLIT

Both SDKs extend OpenTelemetry's capabilities to capture LLM-specific data.

Feature Comparison: Traceloop vs OpenLIT

FeatureTraceloopOpenLIT
LLM Providers
OpenAI✅✅
Azure OpenAI✅✅
Anthropic✅✅
Cohere✅✅
Ollama✅✅
Mistral AI✅✅
HuggingFace✅✅
AWS Bedrock✅✅
Vertex AI (GCP)✅✅
Google Generative AI (Gemini)✅❌
IBM Watsonx AI✅❌
Together AI✅❌
Aleph Alpha✅❌
GPT4All❌✅
Groq❌✅
ElevenLabs❌✅
Vector Databases
Chroma✅✅
Pinecone✅✅
Qdrant✅✅
Weaviate✅❌
Milvus✅✅
Marqo✅❌
Frameworks
LangChain✅✅
LlamaIndex✅✅
Haystack✅✅
LiteLLM✅✅
Embedchain❌✅
Hardware Support
NVIDIA GPUs❌✅

Getting Started

To start using LLM Observability with Middleware:

  1. Choose either Traceloop or OpenLIT SDK based on your requirements.
  2. Instrument your LLM application following the SDK's documentation.
  3. Configure the SDK to send data to your Middleware instance.
  4. Use Middleware's dashboards to monitor and analyze your LLM application's performance.

For detailed setup instructions, please refer to our specific SDK LLM Observability documentation.

Need assistance or want to learn more about LLM Observability in Middleware? Contact our support team in Slack