GVS5H: Zero-Shot Self-Orchestration with Ledger-Based Control for Improved LLM Coding Performance
cs.MA, cs.AI, cs.CL, cs.SE
Submitted: 2026-08-27
Updated: 2026-09-21
License: http://creativecommons.org/licenses/by/4.0/
The gist: Frontier coding performance is typically attained with large, costly proprietary models.
Terminology
Abstract
Frontier coding performance is typically attained with large, costly proprietary models. We introduce ledger-based zero-shot self-orchestration (GVS5H), a training-free method in which fresh instances of one model decompose problems and coordinate through a shared file system. Across eleven open and closed-weight models on the 100 latest hard LiveCodeBench problems, the method yields as much as 25.6 points improvement, boosting several cheaper models to frontier-level performance. Orchestrated Qwen3.8 Flash Next scores 93.0% against Fable 5's 90.4% at 9% of the cost, while the smaller Qwen3.8-27B reaches 92.4%. Gains are not universal: some models are unchanged or worse. Transcript analysis attributes the gain to decomposition and persistent context. Inference-time organization can reach or exceed frontier coding accuracy at a fraction of the cost on self-hostable weights.
Sources
- Large Language Model based Multi-Agents: A Survey of Progress and Challenges
- Rethinking the Bounds of LLM Reasoning: Are Multi-Agent Discussions the Key?
- Single-Agent LLMs Outperform Multi-Agent Systems on Multi-Hop Reasoning Under Equal Thinking Token Budgets
- ReAct: Synergizing Reasoning and Acting in Language Models
- Reflexion: Language Agents with Verbal Reinforcement Learning
- Self-Refine: Iterative Refinement with Self-Feedback
- Sakana Fugu Technical Report
- Learning to Orchestrate Agents in Natural Language with the Conductor
- RouteLLM: Learning to Route LLMs with Preference Data
- MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework
- ChatDev: Communicative Agents for Software Development
- AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation
- CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Society
- AgentCoder: Multi-Agent-based Code Generation with Iterative Testing and Optimisation
- MapCoder: Multi-Agent Code Generation for Competitive Problem Solving
- TRINITY: An Evolved LLM Coordinator
- ARIADNE: Agentic Reward-Informed Adaptive Decision Exploration via Blackboard-Driven MCTS for Competitive Program Generation
- Exploring Advanced LLM Multi-Agent Systems Based on Blackboard Architecture
- Improving Factuality and Reasoning in Language Models through Multiagent Debate
- More Agents Is All You Need
Related papers
- Highway Congestion Reduction through Reinforcement Learning Based Eulerian Headway Control
- Temporal Fair Division in Multi-Agent Systems: From Precise Alternation Metrics to Scalable Coordination Proxies
- Auditable and Deployable Multi-Agent Motion Planning via Posetal Games from Legal Guidelines
- ORACLE-SWE: Quantifying the Contribution of Critical Information Signals on SWE Agents
- You Only Align Once: Propagating Cooperative Behaviors in Multi-Agent Systems through Seed Agents
- Deny Without Disabling: Authorization-Paired Evaluation and Control for Multi-Agent Systems