Skip to main content

​
Roadmap to 0.2.0

  • Improving/reinventing the testing strategy
  • Guardrails hardening: better UI, simpler architecture, easier custom guardrails, and response-side guardrails applied before output reaches the client
  • Writing the customized Code of Conduct
  • Plugins - at least for guardrails
  • OpenTelemetry (OTel) support
  • Full support for the OpenAI /conversations lifecycle
  • Full support for the OpenAI /responses lifecycle
  • Anthropic-compatible /messages ingress and /messages/count_tokens
  • Budget management with limits per user_path and/or API key
  • Editable model pricing for accurate cost tracking and budgeting
  • Prompt cache visibility showing how much of each prompt was cached by the provider
  • UI visibility for prompt caching and local cache usage
  • Fixed failover charts in the dashboard

​
Roadmap to 0.3.0

  • Hardening and extending MCP proxy features
  • Improving the overall quality and simplicity of configuration and UI
  • Rethinking and reworking pass-through API
  • Support for Gemini native API and translations from Gemini native API calls to other dialects
  • Default prices correctness improvements
    • setting up an AI curator / agent to monitor the prices across the providers websites
  • Improving UI and Audit Logs view performance
  • Improving resilient features (explicit failure + failover for cases like vLLM stopped in the middle of streaming etc.)
  • Facilitate migration from LiteLLM to GoModel (script/command + config.yaml convertedr + guide)
  • Better support for /audio related endpoints and modalities
  • Better for Video Generation (storing artifacts and more)
  • Provider-native passthrough for all providers, beyond the current beta coverage
  • Failover for the image endpoints, so image generation and edits can fall back across providers like chat does
  • Streaming image generation (SSE relay of partial images on /v1/images/*)
  • Plugins hardening
  • Driving model properties by models.json file attributes more than the model names
    • <think> / <thinking> blocks instead of a separated field in the response
    • Supported endpoints per model
  • Dashboard customization (more charts, more data to surface, waterfall charts)
  • Workflows rework with better UI/UX and a more flexible design

​
Commercial features (Pro/Enterprise version)

  • Intelligent routing (GoModel Pro, beta)
  • Context window compression (GoModel Pro)
  • OIDC single sign-on (GoModel Pro)
  • Cluster mode
  • LDAP integration
  • Role-based access control (RBAC) for the dashboard and admin API
  • Secret manager integrations (Vault, AWS Secrets Manager/KMS, Azure Key Vault, GCP Secret Manager)
  • Managed outbound proxy pools: health checks, failover between proxies, and assignment rules across providers (per-provider proxy_url is in core)
  • Audit-log export/streaming to object storage and data lakes
  • IP / CIDR allowlists
  • LTS channel with signed artifacts, SBOM, and CVE pre-notification
  • SAML 2.0 + SCIM directory sync

​
Any suggestions?

The roadmap is shaped by what users actually need - tell us what’s missing:
Last modified on September 22, 2026