GraphWorkflow paper: compile-once engine with proofs, cost model, and open benchmarks
KyeGomezB · x · 2026-09-08
The Swarms team published the full systems paper behind GraphWorkflow (Kye Gomez et al.): a formal execution model with proofs, a cost model for agent orchestration, engine design details, and an open benchmark suite against LangGraph 1.0.4 (5 topologies, 10–200 nodes, 15 configurations).
- Compiled-graph execution: 7.0x geometric-mean speedup, 62.5x on 200-node chains
- 21.6x–31.3x faster compilation; 7.9x faster cold path
- All numbers are medians of 9 samples with 95% CIs; harness, raw data, and reproduction scripts are open
- A companion 14-minute blog post walks through the cost taxonomy, compile-once architecture, and how to reproduce every number
More from coding & agent
- BELL: Open-Source Stack Combines Browserbase, Eve, iMessage and Stripe Link for Personal Agents — jeff_weinstein · 2026-10-27
- Legora runs 2,000 parallel agent evals on Modal Sandboxes for legal AI — blelbach · 2026-10-27
- Two engineers rebuilt a FedEx supplier's billing core in 3.5 months vs an estimated 7-8 — alex_verem · 2026-10-27
- GPT-6.1 Terra gains ~20 points on Terminal-Bench 4.0, rivalling GPT-6.1 Sol — ArtificialAnlys · 2026-10-27
- GPT-6.1 Terra scores 48 on AA Coding Agent Index at one-third the cost of GPT-5.6 Terra — ArtificialAnlys · 2026-10-27
- Shift-1 router cuts coding agent spend up to 55% while keeping SWE-bench accuracy — ypatil125 · 2026-10-27