Compumacy/Psych_medgemma

TEXT GENERATIONConcurrency Cost:2Model Size:27BQuant:FP8Ctx Length:32kPublished:Jul 3, 2025License:apache-2.0Architecture:Transformer0.0K Open Weights Cold

Compumacy/Psych_medgemma is a 27 billion parameter experimental large language model developed by Daemontatox, fine-tuned from medgemma-27b with a 32768 token context length. It specializes in clinical psychology and psychiatry, designed to assist mental health professionals by processing complex clinical vignettes and generating structured, evidence-based responses aligned with diagnostic manuals like DSM-5-TR and ICD-11. This model excels at systematic differential diagnosis, comprehensive risk assessment, and evidence-based treatment recommendations.

Loading preview...

Compumacy/Psych_medgemma: Specialized AI for Clinical Psychology & Psychiatry

Compumacy/Psych_medgemma is an advanced, experimental 27 billion parameter large language model, fine-tuned by Daemontatox from the medgemma-27b base. It is specifically designed to augment mental health professionals in clinical assessment and treatment planning, leveraging a 32768 token context length. The model systematically applies diagnostic criteria from the DSM-5-TR, references ICD-11 classifications, and cites peer-reviewed literature to support its recommendations.

Key Capabilities

  • Chief Complaint Analysis: Objectively summarizes presenting symptoms.
  • Differential Diagnosis: Systematically evaluates potential diagnoses against DSM-5-TR criteria.
  • Comprehensive Risk Assessment: Identifies risks related to suicide, homicide, psychosis, and substance use.
  • Evidence-Based Recommendations: Proposes treatment plans (pharmacotherapy, psychotherapy) based on guidelines from APA, WFSBP, and NICE.
  • Monitoring and Referrals: Outlines necessary follow-ups and indications for specialized care.

Training and Performance

The model was fine-tuned using the Unsloth framework for high-performance and memory efficiency on the Daemontatox/Psy-Data-books dataset, which comprises professional psychiatric literature, textbooks, and clinical guidelines. This enables it to generate responses dense with relevant, domain-specific information, emulating the structured reasoning process of a psychiatrist.

Ethical Considerations

It is crucial to note that this model is a research-focused tool and not a substitute for a qualified medical or mental health professional. All outputs require independent verification by a licensed clinician. Users must avoid inputting Protected Health Information (PHI) or Personally Identifiable Information (PII).