Agentic Workflow Boosts Small LLM Medical Diagnosis Accuracy

Mahmood Bayeshi, Veysel Kocaman, Muhammed Ali Naqvi, Yigit Gul, David Talby· July 28, 2026 View original

Summary

The DeepLens Diagnosis Agent, a five-stage agentic workflow, enables a small medical reasoning model (JSL Medical Small 7B v2) to achieve 60.14% top-1 diagnostic accuracy on the DiagnosisArena benchmark. This performance surpasses larger frontier LLMs and demonstrates that disciplined workflow design can significantly enhance diagnostic reasoning and cost-efficiency.

A new research paper introduces the DeepLens Diagnosis Agent, a sophisticated five-stage agentic workflow designed to significantly improve medical diagnostic reasoning using smaller language models. While frontier LLMs are powerful, single-shot prompting often leads to brittle diagnostic outcomes. This agentic pipeline, centered on a 7B-parameter medical reasoning model and Retrieval-Augmented Generation (RAG), enforces a structured process. The workflow includes disciplined clinical extraction, targeted retrieval of knowledge, constrained candidate generation, explicit evidence triangulation, and an auditable final decision. This structured approach yielded a 60.14% top-1 diagnostic accuracy on the 915-case DiagnosisArena benchmark, a remarkable 36-point gain over the same model without the agentic workflow. This performance not only makes it the highest among small and medium-sized models but also allows it to outperform larger models like Claude Sonnet 4.5 and Gemini 3.1 Pro. Beyond its superior accuracy, the DeepLens Diagnosis Agent offers significant cost savings, operating at USD 0.0072 per case, making it 35-45% cheaper than the frontier models it outperforms. The pipeline also generates structured intermediate artifacts, enhancing traceability, reproducibility, and error localization, which are critical for high-stakes applications like medical diagnosis. This research underscores that intelligent workflow design can outweigh raw model size or API cost in achieving superior and auditable results.

Why it matters

This research demonstrates that strategic workflow design can enable smaller, more cost-effective AI models to achieve performance comparable to or better than larger, more expensive frontier models, especially in critical domains like medical diagnosis where accuracy and auditability are paramount.

How to implement this in your domain

  1. 1Analyze existing LLM applications to identify areas where structured agentic workflows could improve performance and reliability.
  2. 2Design multi-stage pipelines for complex tasks, breaking them down into discrete, auditable steps.
  3. 3Experiment with smaller, specialized models combined with RAG and agentic orchestration for cost-effective solutions.
  4. 4Prioritize the generation of intermediate artifacts for transparency and error localization in AI-driven processes.

Who benefits

HealthcarePharmaceuticalsLegalCustomer ServiceFinance

Key takeaways

  • Agentic workflow design can dramatically improve the performance of smaller LLMs, even surpassing frontier models.
  • Structured multi-stage pipelines enhance diagnostic reasoning, accuracy, and auditability.
  • This approach offers significant cost savings compared to relying solely on large, expensive LLMs.
  • Traceability and error localization are crucial benefits for high-stakes AI applications.

Original post by Mahmood Bayeshi, Veysel Kocaman, Muhammed Ali Naqvi, Yigit Gul, David Talby

"arXiv:2607.22555v1 Announce Type: new Abstract: Medical diagnosis is a multi-stage process: extract facts, consult knowledge, generate a differential analysis, and select the best diagnosis with explanations. Frontier LLMs are strong generalists, but single-shot prompting often y…"

View on X

Originally posted by Mahmood Bayeshi, Veysel Kocaman, Muhammed Ali Naqvi, Yigit Gul, David Talby on X · view source

Want to go deeper?

Turn these trends into skills with Learnijoy's hands-on AI & tech courses.

Explore courses