Skip to main content
Ctrl+K
nvMolKit - Home nvMolKit - Home

nvMolKit

nvMolKit - Home nvMolKit - Home

nvMolKit

Table of Contents

  • Overview
  • Molecular similarity on GPU
  • Force field optimization on GPU
  • Hardware autotuning
  • Cursor / agent skill
  • nvMolKit APIs
    • MorganFingerprintGenerator
    • pack_fingerprint
    • unpack_fingerprint
    • crossTanimotoSimilarity
    • crossTanimotoSimilarityMemoryConstrained
    • crossCosineSimilarity
    • crossCosineSimilarityMemoryConstrained
    • EmbedMolecules
    • MMFFOptimizeMoleculesConfs
    • MMFFBatchedForcefield
    • MMFFBatchElement
    • UFFOptimizeMoleculesConfs
    • UFFBatchedForcefield
    • UFFBatchElement
    • butina
    • hasSubstructMatch
    • countSubstructMatches
    • getSubstructMatches
    • SubstructSearchConfig
    • SubstructMatchResults
    • GetConformerRMSMatrix
    • GetConformerRMSMatrixBatch
    • GetTFDMatrix
    • GetTFDMatrices
    • AsyncGpuResult
    • HardwareOptions
    • is_available
    • tune_embed_molecules
    • tune_mmff_optimize
    • tune_uff_optimize
    • tune_batched_forcefield
    • tune_substructure
    • save
    • load
    • TuneResult
  • Changelog
  • nvMolKit APIs
  • save

save#

nvmolkit.autotune.save(
config: HardwareOptions | SubstructSearchConfig,
path: str | PathLike,
) → None#

Persist config to a JSON file at path.

The serialized payload includes a type tag so load() can return the correct class without the caller specifying it.

previous

tune_substructure

next

load

On this page
  • save()
NVIDIA NVIDIA
Privacy Policy | Manage My Privacy | Do Not Sell or Share My Data | Terms of Service | Accessibility | Corporate Policies | Product Security | Contact

Copyright © 2025-2026, NVIDIA Corporation & Affiliates.