DockingAutoDock Vina

AutoDock Vina vs Glide vs GOLD: Molecular Docking Software Compared

Compare AutoDock Vina, Schrodinger Glide, and CCDC GOLD across accuracy, speed, licensing, and cost. Find the best docking tool for your research.

Ryan Bethencourt
March 24, 2026
8 min read

Three Major Docking Programs

Molecular docking software predicts how small molecules bind to protein targets. Three programs dominate the field: AutoDock Vina (free, open-source), Glide (Schrodinger, commercial), and GOLD (CCDC, commercial). Each uses a different scoring function and search algorithm, and each has distinct strengths. This comparison covers the practical differences that matter when choosing a tool for your research.

AutoDock Vina: Open-Source Standard

Vina was released in 2010 and has become the default docking tool in academic research. It uses an empirical scoring function trained on experimental binding data, combined with an iterated local search algorithm for pose optimization. Vina is fast, free, and well-suited for high-throughput screening.

  • Scoring: empirical function combining van der Waals, H-bonds, hydrophobic, and torsional terms
  • Search: iterated local search with random restarts (Broyden-Fletcher-Goldfarb-Shanno optimizer)
  • Speed: 1 to 3 minutes per ligand on a modern CPU
  • License: Apache 2.0 (free for all use)
  • Strengths: speed, ease of use, massive community, API availability via SciRouter
  • Weaknesses: limited handling of water molecules and metal coordination

Glide: Industry Standard from Schrodinger

Glide is the docking component of the Schrodinger computational chemistry suite. It offers three accuracy tiers: HTVS (high-throughput virtual screening), SP (standard precision), and XP (extra precision). Glide is the most widely used commercial docking tool in pharmaceutical industry pipelines.

  • Scoring: GlideScore empirical function with penalties for desolvation, strain, and rotatable bonds
  • Search: hierarchical funnel: initial rough positioning, then Monte Carlo sampling, then energy minimization
  • Speed: HTVS is comparable to Vina; XP is 10 to 50 times slower
  • License: commercial ($15,000 to $50,000+ per year)
  • Strengths: XP mode accuracy, integrated with Schrodinger ecosystem (FEP+, ADMET, Maestro)
  • Weaknesses: expensive, requires Maestro GUI or Schrodinger Python API

GOLD: Genetic Algorithm Approach from CCDC

GOLD (Genetic Optimisation for Ligand Docking) uses a genetic algorithm to explore conformational space. It is developed by the Cambridge Crystallographic Data Centre and is particularly strong at handling ligand and receptor flexibility. GOLD offers multiple scoring functions that can be selected based on the use case.

  • Scoring: multiple options including GoldScore (force-field based), ChemScore (empirical), ASP (knowledge-based), and PLP (piecewise linear potential)
  • Search: genetic algorithm with crossover and mutation operators
  • Speed: 5 to 15 minutes per ligand (slower than Vina and Glide SP)
  • License: commercial ($5,000+ per year academic, higher for commercial)
  • Strengths: excellent flexible docking, multiple scoring options, strong for pose prediction
  • Weaknesses: slower, less suited for large-scale screening

Head-to-Head Comparison

Accuracy

On the CASF-2016 benchmark (a standardized set of 285 protein-ligand complexes), Glide XP and GOLD with ChemScore tend to outperform Vina in both pose prediction (RMSD to crystal pose) and scoring correlation with experimental binding affinities. However, the differences are often small, and all three tools correctly predict binding poses within 2 angstroms RMSD for 60 to 80 percent of targets.

Speed and Throughput

Vina is the fastest for single-ligand docking and scales well for screening campaigns. Glide HTVS mode matches Vina in speed but requires a commercial license. GOLD is the slowest of the three but compensates with more thorough conformational sampling.

Cost and Accessibility

This is where the three tools diverge sharply. Vina is free and can be accessed via API through SciRouter, requiring no local installation or license. Glide and GOLD require institutional licenses that can cost tens of thousands of dollars per year, limiting access for smaller labs, startups, and individual researchers.

Note
The cost barrier of commercial docking software is one of the primary reasons Vina has dominated academic research. API access through SciRouter further reduces the barrier by eliminating installation and infrastructure requirements.

Using Vina via SciRouter API

While Glide and GOLD require local installations with valid licenses, Vina is available as a cloud API through SciRouter. This means you can integrate molecular docking into any workflow with a simple HTTP call:

Quick Vina docking via SciRouter
import requests

response = requests.post(
    "https://api.scirouter.ai/v1/docking/vina",
    headers={"Authorization": "Bearer sk-sci-your-api-key"},
    json={
        "ligand_smiles": "C1=CC=C(C=C1)C(=O)O",  # Benzoic acid
        "receptor_pdb": "1FJS",
        "exhaustiveness": 8
    }
)
print(response.json()["job_id"])

Which Should You Choose?

The right tool depends on your context:

  • Academic screening on a budget: Vina via SciRouter API. Free, fast, no installation.
  • Pharma lead optimization pipeline: Glide XP for accuracy, integrated with Schrodinger FEP+.
  • Flexible receptor docking: GOLD with its genetic algorithm handles receptor side-chain flexibility well.
  • AI-augmented workflows: Combine Vina with DiffDock to compare physics-based and AI-based docking results.

Try AutoDock Vina from the SciRouter tools page. For a step-by-step walkthrough, see our AutoDock Vina beginner tutorial.

Frequently Asked Questions

Which docking software is most accurate?

On standardized benchmarks like the CASF-2016 dataset, Glide XP tends to show the best correlation with experimental binding affinities, followed by GOLD and then Vina. However, accuracy varies significantly by target class. For kinases and GPCRs, all three perform similarly. The best approach is to benchmark on your specific target if accuracy is critical.

Which docking software is cheapest?

AutoDock Vina is free and open-source under the Apache 2.0 license. GOLD requires a CCDC license starting around $5,000 per year for academic users and significantly more for commercial use. Glide is part of the Schrodinger Suite, which typically costs $15,000 to $50,000+ per year depending on the license tier. SciRouter provides Vina via API with free credits to get started.

Can I use Glide or GOLD via an API?

Glide and GOLD do not offer public cloud APIs. They require local installation with valid licenses. Schrodinger offers Schrodinger Cloud for enterprise customers, but it is not a public API. SciRouter currently provides AutoDock Vina and DiffDock as API-accessible docking tools.

Which is best for virtual screening?

For large-scale virtual screening (100,000+ compounds), Vina is the most practical choice due to its speed, parallelizability, and zero licensing cost. Glide SP mode is also widely used in industry for virtual screening but requires a commercial license. GOLD is generally slower and better suited for smaller-scale rescoring or lead optimization.

Try It Free

No Login Required

Try this yourself

500 free credits. No credit card required.