Swargs Open-Sources GraphWorkflow, a Compile-Once Graph Execution Engine Up to 62.5x Faster Than LangGraph

On September 8, the Swargs team (Kye Gomez et al.) released GraphWorkflow, an open-source multi-agent orchestration engine, along with a full system paper, "GraphWorkflow: A Low-Overhead Compile-Once Graph Execution Engine for Multi-Agent Systems." The engine adopts a "compile once, sweep many" architecture to reduce overhead in large-scale DAG orchestration. Notably, its public benchmarks show execution performance significantly exceeding LangGraph, with gains growing as graph size increases.

Confirmed

• In public benchmarks across 15 topology/scale configurations comparing GraphWorkflow with LangGraph, compiled graph execution achieved a geometric mean speedup of 7.0×, peaking at 62.5× on a 200-node chain
• The paper includes proofs of the execution model, complete benchmarks, and reproduction data; the benchmarks are open source
• The engine supports multi-agent topologies such as parallelism, branching, fan-in/fan-out, loops, and complex coordination expressed as DAGs

Why it matters

• The team explains the motivation: multi-agent systems need expressiveness beyond simple pipelines; while LangGraph established this direction, the team argues its syntax is complex and its orchestration layer underperforms. GraphWorkflow's results show that orchestration overhead grows with graph size, and compiled execution can substantially cut that cost
• With reproducible benchmarks and the paper, it offers the community an open-source orchestration option independent of LangGraph, and a reference for quantifying orchestration-layer overhead

2026-09-08 ~ 2026-09-08 · 10 related posts

• [source] Swargs unveils GraphWorkflow: 7x faster than LangGraph, up to 62.5x on 200-node chains — KyeGomezB · 2026-09-08
• Sarms paper claims GraphWorkflow runs compiled agent graphs up to 62.5x faster than LangGraph — KyeGomezB · 2026-09-08
• Why Swarms rebuilt multi-agent orchestration: LangGraph's syntax and performance limits — KyeGomezB · 2026-09-08
• [source] Bigger graphs, bigger gains: GraphWorkflow beats LangGraph across 15 configurations — KyeGomezB · 2026-09-08
• DAG agent orchestration with GraphWorkflow: same workflow in half the code — KyeGomezB · 2026-09-08
• Inside GraphWorkflow: Kahn's algorithm, frozen plans, and runtime overhead near zero — KyeGomezB · 2026-09-08
• [source] GraphWorkflow full paper now available with open benchmark data — KyeGomezB · 2026-09-08
• GraphWorkflow paper: compile-once engine with proofs, cost model, and open benchmarks — KyeGomezB · 2026-09-08
• GraphWorkflow resource roundup: paper, blog, GitHub, and docs — KyeGomezB · 2026-09-08
• Swarms releases GraphWorkflow, 7x faster than LangGraph for multi-agent graphs — KyeGomezB · 2026-09-08