ARBR vs Pencil: Features, Pricing & Which Is Better (2026)
A side-by-side comparison of ARBR and Pencil — 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.
Pencil
Pencil (Pencil Works)
In-IDE design canvas that lets engineers design UI and convert designs directly into code to speed development.
Key features
- In-IDE Design Canvas: Provides a visual design canvas embedded inside developers' IDEs so users can create and edit UI designs without context switching to external design apps.
- Design-to-Code Export: Converts designs into code-friendly artifacts or components (land in code) to reduce manual implementation and speed delivery of production UI.
- Local MCP Server & Discovery: Runs a local MCP-style service that writes config (e.g., ~/.claude.json) so external tools and agents can discover Pencil, read its port/command, and integrate programmatically.
- Auto-Configuration & Port Sync: Supports auto-setup and auto-sync behaviors (project or global scope) so integrations (like OpenCode) automatically add Pencil to their configs and update ports when Pencil restarts.
- Project-Scoped Integration: Can be configured per-project or globally, allowing teams to make Pencil available only within specific repositories or across all projects for consistent workflows.
- Tooling & CLI Support: Exposes status, setup, and sync actions for partner tools and agent plugins (e.g., pencil_status, pencil_setup, pencil_sync) to manage connectivity and lifecycle programmatically.
- Canvas-based UI design embedded into developer workflows/IDEs
- Design-to-code export so visual designs can be converted into code artifacts
- Runs a local MCP/config server and writes runtime config to a file (e.g., ~/.claude.json) for tooling integrations
- Integration plugins (example: OpenCode pencil-sync) that auto-setup Pencil in project/global configs and auto-sync ports after restarts
- CLI/tools for status, setup and sync (pencil_status, pencil_setup, pencil_sync) to manage Pencil integration lifecycle
- Support for project or global scope configuration (config added to ~/.config/opencode/opencode.jsonc or ./opencode.json)
- Automatic port/config backup and validation when updating third-party tool configs
Best for
- IDE-native UI Implementation: Designers and engineers create UI layouts directly inside an IDE and export components to the codebase, eliminating design handoffs and manual translation.
- Agent-enabled Code Generation: Integrate with developer agents or automation (via Pencil's MCP config) to let bots generate, update, or wire up UI code from canvas designs automatically.
- Project-scoped Tooling Integration: Add Pencil to a single project's config to enable team members to access the same design-canvas tooling and keep design artifacts synchronized with that repository.
- Auto-recovery After Restarts: Use automatic port-sync and setup tools so external tooling (like OpenCode) recovers connectivity after Pencil restarts without manual reconfiguration.
- Rapid Prototyping and Iteration: Quickly prototype UI ideas inside the IDE, iterate in-context with code, and validate implementations faster by seeing designs translate to code artifacts immediately.
- Cross-tool Workflow Automation: Connect Pencil with CI, linters, or component libraries via its MCP endpoint to automate tasks like component registration, export, or style enforcement from designs.
- Embed visual design work directly inside developers' IDEs to reduce context switching
- Automatically export or convert canvas designs into code components or project assets
- Auto-configure developer tools and agents to use the locally running Pencil service
- Keep IDE/tooling configurations in sync when Pencil restarts or changes port
- Integrate Pencil with automation agents or local development frameworks to enable programmatic access to designs