> ## Documentation Index
> Fetch the complete documentation index at: https://arize-ax.mintlify.site/docs/llms.txt
> Use this file to discover all available pages before exploring further.
# Integrations
export const NpxEvalsHero = ({children, footer}) =>
;
## Instrument in one command
Not sure which integration you need? Run this from your terminal to launch your coding agent, sign up for a free Arize AX account, and have it detect your stack and wire up the right integrations below — both your framework and the provider it wraps.
## Request or contribute an integration
Don't see an integration you were looking for? We'd love to [hear from you!](https://github.com/Arize-ai/openinference/issues/new/choose)
## Integration types
Arize has a wide range of integrations that fall into into the following categories:
1. **Tracing integrations**: Arize AX captures application traces from applications built using a specific library.
2. **Evaluation integrations**: For evaluation integrations, you can use our **python eval library** to make calls to a specific model, or you can use a **third-party eval library** to evaluate Arize traces and visualize in Arize AX.
Each partner listing in this section contains **integration docs** and **relevant tutorials.**
## Browse all integrations
The setup pattern is the same across every tracing integration: register a tracer provider with `arize-otel` (Python) / `@arizeai/openinference-*` (TS/JS), then attach the matching instrumentor. Pick your stack below. **When you use an agent framework with an LLM provider, instrument both** — the framework instrumentor does not trace the underlying LLM calls.
### LLM providers
### Agent frameworks
### Coding agents
### OpenTelemetry & SDKs
### Platforms
### Orchestration
### LLM-as-a-Judge model providers
Use an LLM provider as the judge for evaluations via the arize-phoenix-evals library.
### Eval libraries
Grade Arize AX traces with an evaluation library and log the scores back to your spans. Phoenix Evals is Arize's own open-source eval library — the engine the provider judges above plug into — and it needs no separate server.