Key Features

What you can do

Distributed Tracing

Provides real-time monitoring of LLM request flows and bottlenecks using OpenTelemetry.

Automatic Instrumentation

Supports LLM providers (OpenAI, Anthropic, Groq), AI frameworks (LangChain, LlamaIndex), vector databases (ChromaDB, Pinecone), and GPUs (NVIDIA, AMD) without modifying application code.

AI Model Evaluation

Offers evaluation via UI and SDKs for online and offline prompts as well as end-to-end applications.

Kubernetes Observability

Enables zero-code observability with the OpenLIT Operator for automatic injection into Kubernetes deployments.

Real-time Metrics

Tracks token costs, latency, and hallucination detection metrics in real time.