Features
Elastic Compute on bare metal
Block Storage with encryption at rest
Virtual Networking with IPv4/IPv6, private subnets, firewalls
Network Load Balancer with health checks
Managed PostgreSQL with HA, backups, PITR
Attribute-Based Access Control (ABAC)
GitHub Actions managed runners
Premium GitHub Actions runners with 40% faster builds
Kubernetes support (managed)
Burstable VMs from $0.01/hour
Open source (AGPL v3) with self-hosted option
Multi-provider bare metal (Hetzner, Leaseweb, AWS Bare Metal)
Inference endpoints for AI/LLM workloads
GPU virtualization for Nvidia HGX B200 GPUs
Trace visibility for agent steps (prompts, retrievals, tool calls, outputs)
LLM-as-judge evaluation for relevance, toxicity, quality scoring
Dataset creation from traces for reproducible testing
Experiment management and regression benchmarking
Built-in Prompt IDE for iterative prompt optimization
Self-hosted deployment on local, Docker, Kubernetes
Phoenix Cloud managed hosting option
Vendor-agnostic support for any model/framework
Native OpenTelemetry integration
OpenInference specification for LLM telemetry
Human annotation and automated labeling
Ghost trajectories to simulate alternative agent paths
Eval-as-you-test for early quality feedback
One-click integration with LlamaIndex
