Towards Scalable Customization and Deployment of Multi-Agent Systems for Enterprise Applications
Quick Answer
This paper shows that A unified framework for customizing and deploying multi-agent systems enhances enterprise applications by achieving a 4.48x throughput speedup while maintaining performance and robustness.
Quick Take
The approach combines continual pretraining, supervised fine-tuning, and inference optimization techniques like FP8 quantization to address domain-specific needs and reduce latency costs.
Key Points
- Framework enables rapid domain adaptation for in enterprise settings.
- Achieves 4.48x speedup in throughput while maintaining performance on complex tasks.
- Combines continual pretraining, supervised fine-tuning, and preference optimization.
- Integrates speculative decoding and FP8 quantization for cost-efficient serving.
- Addresses high latency and inference costs in agentic workflows.
Paper Resources
📖 Reader Mode
~2 min readAbstract:Large language model (LLM)-based multi-agent systems demonstrate strong performance on complex reasoning and task execution, enabling broad enterprise applications. However, production deployment remains challenging due to domain-specific customization requirements and high latency and inference costs in agentic workflows. We propose a unified framework for customization and efficient deployment of multi-agent systems in real-world settings. The first stage, Agentic Model Customization, combines continual pretraining, supervised fine-tuning, and preference optimization to adapt a compact model to specialized domains while retaining strong agentic capabilities. The second stage, Inference Optimization, integrates speculative decoding and FP8 quantization with targeted calibration to enable cost-efficient serving with minimal quality loss. Across enterprise workloads, our framework enables rapid domain adaptation and achieves a 4.48x speedup in throughput while maintaining performance and improving robustness on long-tail scenarios.
| Comments: | Preprint |
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2606.18502 [cs.CL] |
| (or arXiv:2606.18502v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2606.18502 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Genta Indra Winata [view email]
[v1]
Tue, 16 Jun 2026 21:30:10 UTC (267 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CL
See more →TriAgent: Divergence-Aware Committees for Cost-Efficient Financial Sentiment Analysis
TriAgent introduces a cost-efficient multi-agent system for financial sentiment analysis, combining VADER, FinBERT, and Qwen2.5. It achieves an F1 score of ~0.87 with significant savings of $9.3M/year at a 10M-user scale compared to GPT-4o-mini, while also detecting hallucinations with an AUC of 0.90.