Open-source biomedical deep research assistant powered by Valyu.
Synthesize literature, clinical trials, and FDA drug labels through natural conversation.
Bio is an open-source research assistant built for life sciences professionals who need to run complex biomedical searches across fragmented databases. It automates the search-read-synthesis loop to produce cited reports with deliverables, using Valyu DeepResearch.
It executes prebuilt life sciences workflows across PubMed, ClinicalTrials.gov, FDA drug labels, and patents, trading research depth against turnaround time across fast, standard, and heavy modes. You can inspect a live activity feed of searches and reasoning steps, then export completed reports to PDF or share them with colleagues.
It suits researchers, clinicians, and business development professionals in biotech who spend days stitching together biomedical evidence from disparate databases. It does not suit teams looking for a general-purpose chat interface or those without access to a Valyu API key.
Run typed inputs across pipeline, clinical, regulatory, and business development lenses instead of a blank prompt.
Query PubMed, ClinicalTrials.gov, FDA drug labels, patents, and the web through Valyu's API.
Hover any marker in the generated report to inspect the underlying source title, date, and link.
Generate structured artifacts alongside narratives and export completed reports directly to server-rendered PDFs.
The software is free under the MIT license, but running it requires a Valyu API key which governs usage through its own credit system. Optional Valyu mode adds hosted authentication and billing via Valyu accounts.
880 open remote AI-training roles, updated November 15, 2026; most pay $20–$175 an hour. Zekai lists them free; the application goes to the employer’s platform.
These reviewed AI tools do the same job instantly — from our scored directory.
Bio uses Valyu's unified API to access PubMed articles, ClinicalTrials.gov records, FDA drug labels, patents, and the open web.
Yes, setting NEXT_PUBLIC_APP_MODE=self-hosted enables local SQLite storage, auto-login as a dev user, and removes authentication walls.
Yes, it detects Ollama and LM Studio instances running locally and displays their status and available models in the interface.
See Zekai first in Google