[Roadmap] vLLM Production Stack roadmap for 2025 Q2

This issue document the roadmap for 2025 Q2. We will keep updating this document to include the related issues, pull requests, and discussions in the #production-stack channel in the vLLM slack. Feel free to leave comments / questions in this issue!

Router

Router “frontend”

• (P1) Gateway API extension integration (PR #436 )
• (P2) Router performance enhancements
  • Nuitka compilation for the current router
  • Rust/Go/Nginx-based frontend for router
• (P2) Integration with Application interfaces like MCP or other agentic workflows (#352 )
• (P2) Envoy ext_proc integration (Draft PR: #240)

Router Core Logic

• (P0) Queuing support in the router
• (P0) Prefix aware routing (#239 )
• (P0) KV cache aware routing (#398 )
• (P2) Priority routing
• (P2) Routing to external providers like OpenAI or Anthropic
• (P2) Request migration when the vLLM instance fails
• (P2) Router extension modules
  • Semantic caching (#210 )
  • PII detection (#235 )

Integration with other ecosystem projects

• (P0) Compatibility with vLLM v1
• (P0) Integration with LMCache KV cache controller

Multi-node support

• (P0) Single-vLLM instance multi-node setup (e.g., PP on 2 different nodes) (#422 )
• (P1) The reference implementation for disaggregated prefill (https://github.com/vllm-project/production-stack/pull/340)

K8s-native control plane

• (P0) Model CRD for vLLM deployment operations (#415 )
• (P1) Router CRD for router deployment operations (#415 )
• (P1) LoRA CRD & controller (Proposal: lora-k8s-support.md )
• (P1) Autoscale CRD & controller (Proposal: #238 )

CI/CD and misc.

• (P1) Comprehensive unit tests for router
• (P1) Use CPU-based vLLM for functionality tests (#342 )
• (P1) Github actions for router performance benchmarking (#562 )
• (P2) Github actions for building router docker images for different architectures
• (P2) Release bot to automatically release new versions (helm chart + k8s controller packages + docker images) (#450 )
• (P1) Documentation (#291 )
• (P2) Tutorials for more cloud platforms, different models, and features
• (P2) End-to-end performance benchmarking: (more workload, more setups)

Start with this issue's 2025 Q2 roadmap sections, then follow the linked issues, pull requests, proposals, and the #production-stack Slack discussion for the relevant workstream. There is no single file, test, or entry point named, and completion is not defined beyond updating the roadmap as related work progresses.