ARBR vs Graphiti: Features, Pricing & Which Is Better (2026)
A side-by-side comparison of ARBR and Graphiti — features, pricing, and ideal use cases — to help you decide which AI tool fits your workflow.
ARBR
Gyde & Domkundwar Foundation
Open-source, MIT-licensed AI gateway and control plane that routes, governs and observes every LLM request behind one OpenAI-compatible endpoint.
Key features
- OpenAI-Compatible Routing: A single drop-in endpoint over every major provider, with rules, difficulty-aware selection, cost guardrails and automatic fallback choosing the model per request.
- In-Path Governance: Budgets, rate limits, output guardrails, prompt-injection checks and kill switches enforce policy before inference rather than auditing it afterwards.
- Structured Observability: Cost, latency, tokens and routing decisions are emitted as structured events attributed by application, team, model and user, viewable in local dashboards or exported to OpenTelemetry backends such as Datadog, Grafana and Prometheus.
- LLM-Judge Evaluation: A sample of live traffic is scored for quality so requests can be routed to the cheapest model that provably clears the bar, rather than optimising on price alone.
- Safe Model Deployment: Canary and shadow new models against real traffic with regression gates that block promotion until evaluations pass, plus instant rollback.
- Broad Provider Coverage: One layer over Anthropic, OpenAI, Google Gemini, Amazon Bedrock, Azure OpenAI, Vertex AI, Groq, DeepSeek, Moonshot, xAI and Mistral, plus LiteLLM and NVIDIA NIM, with pricing and benchmark data for over 3,000 models.
- Drop-In SDK Compatibility: Change only the base URL and existing OpenAI SDKs, agent frameworks and chat UIs keep working, gaining streaming chat completions, embeddings, a realtime voice proxy and JavaScript and Python SDKs.
- Self-Hosted and MIT Licensed: The full control plane runs inside your own infrastructure under an MIT licence, with a hosted option available for teams that do not want to operate it.
Best for
- LLM Cost Reduction: Route summarisation and extraction traffic to cheap small models while reserving frontier models for analysis, cutting spend without hand-editing every call site.
- AI Spend Attribution: Give finance and engineering a per-application, per-team and per-user breakdown of token spend so AI budgets can be owned by the groups that generate them.
- Enterprise AI Governance: Enforce departmental budgets, rate limits and kill switches in the request path so a runaway agent cannot exhaust a quarter's inference budget.
- Provider Risk Mitigation: Keep applications provider-neutral behind one endpoint with automatic fallback, so a single vendor outage or price change does not require a code change.
- Model Migration Testing: Shadow or canary a newly released model against production traffic and let regression gates decide whether it is promoted.
- Prompt-Injection Defence: Apply output guardrails and prompt-injection checks centrally for every application instead of reimplementing them per service.
Graphiti
getzep (GitHub)
Open-source project to build real-time knowledge graphs and persistent memory stores for AI agents.
Key features
- Real-time Graph Construction: Extracts entities and relationships from incoming text and builds a dynamic graph representation so agent context is stored as structured nodes and edges for fast retrieval.
- MCP-Compatible Server API: Exposes endpoints and protocols aligned with Model Context Protocol patterns to let AI agents query episodes, entities, and contextual graph data as persistent memory.
- Document Ingestion and Registration: Registers documents across multiple formats into the graph store, enabling documents to be linked, searched, and referenced by agents in retrieval workflows.
- Graph Database Integration: Supports integration with graph-backed storage (examples and forks reference Neo4j and FalkorDB) to persist entities, relationships, vectors, and perform graph queries.
- Episode-Based Memory Management: Groups interactions into episodes with metadata (UUIDs, timestamps) to enable chronological context, session tracking, and selective retrieval of past interactions.
- Multi‑Project & Docker Deployment: Community forks and examples provide CLI and Docker Compose setups to run root and project-specific MCP servers, enabling multi-project sharing of a single graph database.
- Developer Tooling & Extensibility: Source-code-first, open repository structure allows customization, extension, and integration into agent stacks and RAG pipelines.
- Extract entities and relationships from text to build knowledge graphs
- Persist graph data in Neo4j graph database
- Model Context Protocol (MCP) server implementation for context serving
- Docker Compose and CLI tooling for quick local deployment and multi-project setups
- Support for project-specific MCP servers sharing a common database (multi-tenant graphs)
- Document registration and ingestion across multiple file formats for RAG workflows
- Integrations/examples showing usage with Cursor and agent systems to store prompts as graph memory
Best for
- Persistent Conversational Memory: Provide chatbots and assistants with long-term memory by storing and retrieving entities and relationships learned across sessions.
- RAG Backend for Document Search: Index and link documents into a knowledge graph so retrieval-augmented generation pipelines can find relevant passages via graph relationships and metadata.
- Agent Context Sharing Across Projects: Run multi-project MCP servers so multiple agents or teams can share and query a centralized knowledge graph for consistent context.
- Debugging and Traceability: Use episode grouping and entity links to trace agent decisions back to source documents and previous interactions for audit and improvement.
- Entity Relationship Discovery: Extract and visualize relationships across ingested content to discover connected concepts, people, locations, or events for analytics or recommendation systems.
- Providing persistent structured memory for conversational AI agents
- Backend for retrieval-augmented generation (RAG) systems using graph storage
- Indexing and searching entities/relations from ingested documents
- Multi-project knowledge graph deployments that share a central Neo4j instance
- Developer experimentation and prototyping of graph-based context for models