PharmaLLM Agent

PharmaLLM Agent is an open-source medical dialogue agent and MRI segmentation workspace for clinical tasks.

PharmaLLM Agent is a medical dialogue agent and MRI segmentation workspace built on PharmaLLM. It provides features for language model fine-tuning, evaluation, agent tool orchestration, asynchronous image segmentation, and clinical review. Users can upload images, initiate femoral head and acetabulum segmentation using natural language, and view results in a four-view workspace. The system supports inputs like NIfTI and DICOM files, utilizes LangGraph for agent orchestration, and runs language and segmentation models in independent GPU processes. It is designed for medical teams with role-based permissions for doctors and reviewers.

PharmaLLM Agent at a glance

Best for: Researcher, Software engineer
Runs on: self-hosted, open source
Pricing: Free
Released: October 2026
Category: AI agent tool
Website: github.com/xxpcy/pharmallm-agent

Key features

Stream responses and manage conversation history
Orchestrate tools using LangGraph state graphs
Parse NIfTI and DICOM imaging formats
Perform natural language MRI segmentation
Support team collaboration with role-based reviews
Run independent vLLM and nnU-Net services
Manage tenant and user authorization

Use it for

Conduct medical dialogue and Q&A sessions
Segment femoral head and acetabulum from MRI
Review and approve segmentation results
Evaluate language models on medical exams

Alternatives to PharmaLLM Agent

MemPalace — Achieves high recall with local-first storage and zero API calls.
FastGPT — It uses hybrid retrieval and auto-cleansed data to prevent hallucinations.
Limma — It answers queries with complete sets and proof instead of partial search fragments.
SimpleMem — Provides semantic lossless compression to reduce token consumption and maintain long-term context.
OpenRMM — Self-hosted infrastructure with complete data privacy and human approval for every action.
ReMe — It enables local-first, file-native long-term memory sharing across multiple AI agents.
MuscleMemory — It cuts input tokens, tool calls, and execution time while maintaining task success.
MemMachine — Provides long-term persistent memory for LLM-based agents.
MS-Agent — It supports multimodal inputs, autonomous scheduling, and modular tool integration.
Memori — Reduces LLM token usage and spend by over 97% with targeted recall.
Memory OS for Hermes — Stops agents from forgetting by adding local persistent memory layers.
MyAgent — It features built-in permission controls, long-term memory, and skill self-evolution.

All PharmaLLM Agent alternatives