#parallelism

2 posts

vLLM x AgentX: Optimizing for Real-World Agentic Serving

Sep 8, 2026·24 min read

How vLLM optimizes KV cache management, parallelism, scheduling, and P/D disaggregation for agentic workloads, validated on SemiAnalysis AgentX with up to 130K tokens per GPU-second and a 14.6x-106x serving-cost advantage over Opus 5.

Efficient Decode Context Parallelism with vLLM for Long Context Workloads

Aug 7, 2026·12 min read

Decode Context Parallelism (DCP) in vLLM shards KV cache across GPUs by sequence dimension, enabling 3× higher throughput on long-context agentic workloads compared to standard tensor parallelism.