Numbers you can check. Methods you can challenge.

FluxMateria benchmarks include accuracy metrics, comparison methodology, and validation datasets.

Our benchmark philosophy

📋

Publish the methodology

How we test, what datasets we use, how we measure. No hidden assumptions.

⚖️

Show head-to-head comparisons

Against established tools where possible. Fair comparisons, same test sets.

🎯

Document validation scope

Where predictions are most and least reliable. We tell you the boundaries.

🔄

Enable reproduction

Enough detail that you could replicate. Trust but verify.

Key metrics at a glance

Summary of performance across core capabilities.

Bond Length Accuracy

<0.5%error
vs experimental data

DFT-competitive accuracy on molecular geometry across diverse chemical spaces.

Screening Throughput

10,000+mol/hr
full property panel

Single-thread performance. Batch processing scales linearly with cores.

ADMET Panel

253K+
Compounds Tested

Full-scale ADMET validation across PPB, BBB, solubility, metabolism, and toxicity. Dual-presented: curated + large-scale results.

Materials

313
Materials Validated

100% pass rate on band gap predictions across 14 material families. III-V, II-VI, TMDs, perovskites, oxides.

Benchmarks by module

Detailed performance data for each capability.

💊 ADMET

Production
82.6%
CYP Panel Accuracy
81.8%
BBB Accuracy
812-1,035
mol/sec
253K+
Compounds Validated
  • Permeability: r=+0.964, 86.7% accuracy (40,318 compounds) ✓
  • CYP Inhibition: 82.6% overall (5 isoforms, v4.1) ✓
  • BBB: 81.8% accuracy, 87% sensitivity (7,803 compounds) ✓
  • Metabolism: r=+0.753, 70.3% accuracy (38,482 compounds) ✓
  • PPB, hERG, DILI, Solubility — all validated at scale ✓

253K+ compounds tested across 8 ensemble-validated endpoints.

Full Results → Methodology

🔬 Materials

Production
0.9%
Band Gap MAE
313
Materials
100%
Pass Rate
17
Categories
  • Band gap: 0.9% MAE across 313 materials ✓
  • Universal Derivation v3: 21,550+ semiconductors ✓
  • Blind validation (v8.1): 4.2% Debye error, 100% bandgap type ✓
  • Deterministic — same input always gives same output ✓
Full Results → Module Page

📈 Spectroscopy

Production
6.2%
UV-Vis Error
<1%
IR Error
0.3-0.5
NMR MAE (ppm)
50
UV-Vis Molecules
  • UV-Vis: 6.2% mean error, 50 molecules, 6 categories ✓
  • IR: <1% error, 32 NIST molecules validated ✓
  • NMR: 0.3-0.5 ppm MAE, 10 SDBS molecules, 5 nuclei ✓
Full Results → Module Page

⚗️ Mechanism Discovery

100%
Mechanism Accuracy
168K
Combinations
5.72
kJ/mol MAE
1,000,000x
Faster
  • 49/49 experimental test cases (SN1/SN2/E1/E2) ✓
  • 10,000 random physical consistency tests ✓
  • Head-to-head comparison with DFT (B3LYP) ✓
  • Every prediction traceable and reproducible ✓
Full Methodology & Results → Module Page

MechanismOS

Production
100%
GOLD Direct Ea
94.86%
SILVER Arrhenius Ea
100%
Experimental-only rows
2 tiers
Official validation gate
  • GOLD: 154/154 direct measured activation barriers passed ✓
  • SILVER: 1255/1323 Arrhenius-derived barrier checks passed ✓
  • Official experimental source provenance documented per benchmark tier ✓
  • Control surfaces, optimizer, and audit evidence pack validated end-to-end ✓
Full Results → Module Page

🧪 Synthesis Planning

Production
3.1%
Barrier MAE
29/29
Reaction Types
200
Specific Reactions
<50ms
Per Plan
  • 29 reaction-type barriers at 3.1% MAE (100% pass rate) ✓
  • 200 specific reactions at 0.119% MAE (72 exact matches) ✓
  • 15 disconnection SMARTS patterns validated ✓
  • All barriers fully auditable and reproducible ✓
Full Results → Module Page

⚡ Electron Transfer

Production
26/26
Tests Pass
2–3×
Tunneling Enhancement
Literature
Decay constant match
~150ms
Per Pair
  • Marcus rate constants with FLUX tunneling corrections ✓
  • Through-bond decay constant matches literature ranges ✓
  • Normal, activationless, and inverted Marcus regimes ✓
  • All coupling deterministic and traceable ✓
Full Results → Module Page

🧪 Solvation

0.71
logS MAE (ESOL)
1,128
Compounds
642
FreeSolv
  • Solubility: 0.71 logS MAE on ESOL (below 0.8 target) ✓
  • FreeSolv hydration free energy benchmark (642 molecules) ✓
  • Comparison vs SMD and COSMO-RS solvation models ✓
Full Results →

🧬 BioTarget

Beta
0.537
Pearson r (CASF-2016)
91%
MoA Accuracy
1.90
MAE (pKi)
10,065
Targets
  • Binding affinity: Pearson r = 0.537 on CASF-2016 (270 complexes) ✓
  • MoA prediction: 91% accuracy on ChEMBL validation ✓
  • Target identification: AUC 0.980 ✓
  • Selectivity profiling: planned ⏳
Full Results →

Validation scope

Where FluxMateria predictions are most and least reliable.

We document the boundaries of reliable prediction space:

  • ⚠ Novel chemotypes far from validated chemical space
  • ⚠ Specific endpoints with limited experimental data
  • ⚠ Edge cases identified through validation

Confidence indicators in predictions reflect these boundaries. Low confidence = verify experimentally.

Reproducibility

We want you to verify our claims.

Benchmark datasets and evaluation scripts are available to pilot participants.

Request Pilot Access