Version 2.0 Roadmap

Building on the stable v1.0 foundation, v2.0 focuses on developer experience, advanced telemetry, and content-driven insights for the AI community.

Status Legend:

• Done — completed and shipped
• In Progress — actively being worked on
• Planned — scheduled, not yet started

Phase 1: Advanced Telemetry — Done

Operational telemetry with authenticated probes, cost controls, and richer metrics.

• Provider authentication abstraction (Bearer, API key, query param)
• Probe payload normalization (lightweight & reasoning profiles)
• Streaming response validation (transport vs semantic success)
• Per-provider probe policies with daily limits (Redis-backed)
• Probe region tracking (probe_region column + index)
• Database migration for probe region support
• Authenticated latency probes with real inference calls
• Health probe policy integration
• Telemetry freshness badge on provider detail pages
• P50/P95/P99 latency percentile calculations
• Multi-region probe deployment (US, EU, Asia)
• Provider uptime SLA tracking and historical reports

Status: Done

Phase 2: OpenAPI & Swagger — Planned

Full API documentation with interactive exploration.

• OpenAPI 3.1 spec auto-generated from NestJS decorators
• Swagger UI hosted at /api/docs
• Request/response examples for all endpoints
• Authentication documentation (API keys for future use)
• Versioned API schema (v1 namespace)
• SDK code generation from OpenAPI spec

Status: Planned

Phase 3: SDK — Planned

Official TypeScript/JavaScript SDK for programmatic access to the OpenModels API.

• @openmodels/sdk package published to npm
• Type-safe client with full IntelliSense support
• Model listing, search, and detail retrieval
• Provider comparison and ranking queries
• Telemetry data access (health, latency, ranked)
• Pagination helpers and iterator patterns
• Error handling with typed exceptions
• Configurable base URL and timeout
• Browser and Node.js compatibility
• Auto-generated from OpenAPI spec

Status: Planned

Phase 4: CLI — Done

Command-line interface for developers and CI/CD pipelines.

• openmodels-cli package published to npm
• openmodels models list — browse and filter models
• openmodels models info <id> — detailed model info
• openmodels providers list — browse providers
• openmodels compare <model-id> — compare providers for a model
• openmodels search <query> — full-text search
• openmodels telemetry <provider-id> — health and latency data
• Table, JSON, and YAML output formats
• Shell completions (bash, zsh, fish)
• Configuration file (~/.openmodelsrc) for defaults

Status: Done

Phase 5: Insights — Done

Content platform for AI industry analysis, benchmarks, and provider comparisons.

• CMS setup with Payload CMS (headless)
• Insights collection with rich text, taxonomy, and SEO fields
• Article schema: title, slug, lead, cover image, body, tags
• Related models and providers linking
• Draft/published workflow with versioning
• Media management for cover images and inline assets
• Insights listing page on web app
• Individual insight article pages with Lexical rendering
• RSS feed for insights
• Social sharing meta tags (Open Graph, Twitter Cards)

Status: Done

Release Criteria

Before tagging v2.0:

• All telemetry probes running with real authentication
• SDK published to npm with >90% API coverage
• CLI supports all read operations with formatted output
• OpenAPI spec covers 100% of public endpoints
• Insights CMS integrated with web frontend
• Documentation updated for SDK, CLI, and new telemetry features