QBench | Desktop Life Sciences Intelligence
QuantaMCP QBench Workspace Enterprise Documentation Request Access

Desktop Application

Your research. Your machine. No compromises.

QBench brings molecular intelligence to your desktop. 3D visualization, AI-powered analysis, and local computation — with optional cloud connectivity when you need it. Your data never leaves your machine unless you choose.

Request Access

A workspace, not a dashboard

Walk through what it feels like to use QBench, from file drop to insight.

01 — open a molecule

Drag in. See it instantly.

Drag any SMILES string, SDF, MOL, or PDB file into QBench. The 3D viewer renders it instantly with confidence-scored bonds and pharmacophore highlighting.

02 — run analysis locally

Native compute. No setup.

RDKit compute runs natively — no Python environment, no Docker, no cloud dependencies. ADMET predictions, drug-likeness scores, and structural alerts, all on your machine.

03 — ask your AI

Context-aware conversation.

Built-in multi-provider chat (Claude, GPT-4o, Ollama). Ask questions about your molecule in context — the AI sees what you see.

04 — connect when you choose

Cloud on your terms.

Optional QuantaMCP integration gives you cloud intelligence — 122M+ compounds, regulatory compliance, and literature search — without leaving your workspace.

Only in QBench

Ghost Molecule

Ghost Molecule overlay: compare two structures side-by-side with adjustable transparency. See exactly where modifications change the molecular geometry.

  • AI suggests optimized analogs as you work — Tab to accept, Esc to dismiss
  • Three optimization targets pre-computed on every molecule load
  • Adjustable transparency reveals geometric differences at a glance
  • A feature competitors do not have

Choose your depth

QBench works fully offline. Add QuantaMCP when you need cloud-scale intelligence.

QBench alone

No account needed · Offline-capable

  • 3D molecular viewer (NGL)
  • Molecular properties (MW, LogP, TPSA, HBD, HBA)
  • Drug-likeness (5 rule sets)
  • Molecule linter (PAINS + Lipinski)
  • Fingerprint search
  • File explorer + session persistence
  • Batch local properties + CSV export
  • AI chat (bring your own key or use Ollama)

QBench + QuantaMCP

Everything above, plus cloud intelligence

  • ADMET radar + 40+ prediction models
  • Ghost Molecule suggestions
  • FAVES regulatory compliance (DEA, FDA, CWC, EPA, EU REACH)
  • Research explorer (literature, patents, clinical trials)
  • 122M+ compound database
  • Pipeline orchestration
  • Batch ADMET + compliance
  • Credit dashboard

Technical Specifications

Built with Electron + RDKit (native bindings) + Three.js (3D rendering)
Formats SMILES, SDF, MOL, MOL2, PDB, CIF
AI providers Claude, GPT-4o, Ollama (local), custom endpoints
Platform macOS 12+ · Apple Silicon & Intel · 200MB disk
Dependencies None. RDKit is bundled. No Python required.

What's next

Coming soon

Protein Structure Browser

Browse and visualize structures from AlphaFoldDB and PDB directly in QBench. Overlay predicted and experimental structures, inspect per-residue confidence, and export for downstream analysis.

Coming soon

Genomic Data Explorer

Query RefSeq, GenBank, and GEO from your desktop. Explore gene sequences, transcript variants, and expression profiles — connect molecular targets to their genomic context.

In development

Pathway Viewer

Map and explore biological pathways with interactive visualization. Drug targets, gene expression, and protein interactions rendered on pathway networks.

Get early access to QBench

We're onboarding research teams for the desktop beta.

Request Access

v1.0.0 · macOS (Apple Silicon)