DeePMD-kit logo

DeePMD-kit

Start from a pretrained Deep Potential model, fine-tune it for your system, and deploy it at simulation scale.

GitHub release offline packages conda-forge pip install docker pull Documentation Status License

Pretrained models · Fine-tuning · Documentation · Quick start · Model guide · Tutorials · Examples · Releases

[!IMPORTANT] A pretrained model can be your starting point, not just your end result. Download a built-in pretrained DPA4 checkpoint, fine-tune the full model for your system, then test, export, and deploy it through the same DeePMD-kit workflow.

DeePMD-kit turns quantum-mechanical reference data into fast, scalable interatomic potentials. Use it across molecular and materials science—from finite molecules and covalent systems to periodic solids and metals—and scale from laptop fine-tuning to distributed training and MPI-parallel molecular dynamics.

DPA4 model family Pareto frontier for Matbench Discovery CPS and saturated inference throughput

The DPA4 model family traces a Pareto frontier across Matbench Discovery CPS and saturated inference throughput.

⚡ Why DeePMD-kit

Advantage What it unlocks
🧬 Pretrained-first workflows Download pretrained DPA4 models, fine-tune full models, or adapt supported pretrained representations to downstream properties with DPA-ADAPT.
🏗️ Training from scratch Design a model for a new system or physical target, then train it with single-task, multi-task, and distributed workflows across supported backends.
🧠 Modern model portfolio For conservative energy/force interatomic potentials, start with DPA4 for accuracy or DPA4C for simulation throughput and scale.
🧲 More than energy and force Model virials, Hessians, spin and magnetic forces, dipoles, polarizabilities, electronic density of states, atomic populations, and arbitrary intensive or extensive properties.
🔄 Backend flexibility Train or run supported models with TensorFlow, PyTorch, JAX, or Paddle, with backend-aware model formats and conversion paths for compatible architectures.
🚀 Performance from training to MD Use CPUs, CUDA GPUs, ROCm source builds, distributed training, compiled DPA4 paths, compressed DPA4C CUDA inference, AOTInductor .pt2 export, and MPI-enabled simulation.
🔌 Deploy where science happens Use the CLI, Python, C, C++, or Node.js, then connect models to LAMMPS, i-PI, ASE, GROMACS, JAX MD, nvalchemi, OpenMM, Amber, CP2K, ABACUS, and more.
🧩 Open and extensible Compose hybrid potentials, add analytical ZBL or long-range corrections, create custom models and operators, or connect external GNNs such as MACE and NequIP through plugins.

[!TIP] On supported descriptors and workloads, model compression can deliver more than 10× inference speedup and reduce memory usage by as much as 20×. Actual gains depend on the model, system, and hardware.

Backend and interface support varies by model and feature. The web documentation marks compatibility and limitations on each feature page.

🧭 Two starting points, one path to dynamics

flowchart LR
    A["Pretrained DPA4 model"] --> C["Fine-tune on target data"]
    B["Model configuration"] --> D["Train from scratch"]
    E["Target reference data"] --> C
    E --> D
    C --> F["Test, compress, export"]
    D --> F
    F --> G["Python and native APIs"]
    F --> H["Molecular dynamics"]
  1. Choose a starting point: download a pretrained DPA4 checkpoint for adaptation, or configure DPA4 or DPA4C to train from scratch.
  2. Prepare target data in DeePMD's NumPy format or convert structures and trajectories with dpdata.
  3. Fine-tune or train: adapt the full pretrained DPA4 model, or optimize a new DPA4 or DPA4C model with single-task, multi-task, and distributed training workflows.
  4. Validate and export with dp test, dp freeze, backend conversion, embedding extraction, and supported compression paths.
  5. Run simulation through Python or native APIs, or load the model into a supported molecular-dynamics engine.

🚀 Start in minutes

DeePMD-kit requires Python 3.10 or later. The fastest installation path is:

curl -fsSL https://dp1s.deepmodeling.com | bash
dp --version
dp -h

The installation guide covers pip, conda-forge, containers, offline packages, GPU builds, LAMMPS, i-PI, and source installation.

Fine-tune a pretrained DPA4 model

Download a built-in checkpoint, start from its matching released training configuration, and fine-tune it on your target data. This example uses DPA4-Neo, one of the recommended general-purpose sizes:

dp pretrained download DPA4-Neo-OMat24-v20260805
curl -fsSL \
    https://huggingface.co/deepmodelingcommunity/DPA4-OMat24/resolve/main/DPA4-Neo-OMat24-v20260805.json \
    -o input_finetune.json

The DPA4 OMat24 release provides Nano, Mini, Neo, Air, and Plus checkpoints together with their matching training configurations. The downloaded input_finetune.json matches the Neo checkpoint above; for another size or version, use the correspondingly named JSON file. Keep its complete model section unchanged, including the full-periodic-table type_map; replace the training and validation data, and use a smaller learning rate for fine-tuning. Then run:

dp --pt train input_finetune.json \
    --finetune ~/.cache/deepmd/pretrained/models/DPA4-Neo-OMat24-v20260805.pt

These are PyTorch single-task checkpoints, so no model branch selection is needed. They target inorganic materials in the OMat24 chemical space; validate accuracy before using them outside that domain.

The fine-tuning guide covers full-model adaptation. DPA-ADAPT reuses supported pretrained DPA representations for downstream property-prediction tasks.

Pretrained model names can also be resolved and cached automatically by Python:

from deepmd.infer import DeepPot

potential = DeepPot("DPA4-Neo-OMat24-v20260805")

Train a model from scratch

Training from scratch remains a first-class workflow for new architectures, fully custom systems, and physical targets without a suitable pretrained checkpoint. Clone the examples and start with the compact water system:

git clone https://github.com/deepmodeling/deepmd-kit.git
cd deepmd-kit/examples/water/dpa4

# Accuracy-first DPA4 model
dp --pt train input.json

# Or the throughput-first DPA4C model
cd ../dpa4c
dp --pt-expt train input.json

Ready-to-run inputs include:

For a guided end-to-end example, open the web quick-start notebook.

🧠 Choose a model family

For conservative energy/force interatomic potentials, start with the DPA4 family. The choice between its two primary models follows the constraint that matters most for your workload:

Priority Start with Why
Highest accuracy DPA4 SO(3)-equivariant message passing targets the accuracy frontier.
Highest throughput or system scale DPA4C A compact one-hop descriptor targets the throughput frontier and supports compressed CUDA inference.

DPA4 uses the PyTorch backend (dp --pt). DPA4C currently uses the PyTorch Exportable backend (dp --pt-expt); its compressed CUDA path requires float32.

For other physical targets, use the model guide to select a compatible model and backend. The guide also compares data formats, precision, compression, and deployment constraints.

DPA4 and DPA4C energy and force accuracy versus saturated throughput

For energy/force potentials, DPA4 and DPA4C span accuracy–throughput trade-offs for different deployment budgets.

🔬 Go beyond conventional force fields

Goal DeePMD-kit capabilities
Potential-energy surfaces Energy, atomic forces, virials, Hessians, hybrid descriptors, pair tables, and linear model combinations
Magnetic systems Spin-aware descriptors, atomic and magnetic forces, and spin-capable molecular dynamics
Electronic and response properties Dipoles, polarizabilities, density of states, atomic charge populations, and custom property heads
Long- and short-range physics DPLR electrostatics, DPRc range correction for QM/MM, and analytical ZBL bridging
Representation learning Per-atom descriptors, fitting-network features, structural embeddings, clustering, and downstream auxiliary models

Explore the complete set of models and physical targets in the web documentation.

🔌 Deploy into the scientific ecosystem

Inference interfaces

Simulation and workflow integrations

  • LAMMPS, i-PI, ASE, JAX MD, and nvalchemi
  • Ecosystem integrations for OpenMM, Amber, CP2K, GROMACS, ABACUS, DP-GEN, and MLatom
  • External MACE and NequIP models through the DeePMD-GNN plugin

See the integration hub for maintained interfaces, third-party projects, supported scope, and installation guidance.

The native C and C++ interfaces load machine-learning backends as runtime plugins. Applications can therefore open the backend required by a model without directly linking every framework.

[!NOTE] Working with an AI coding or scientific agent? Start with Install with an AI agent, or browse the official Agent Skills for model selection, training, fine-tuning, Python inference, and LAMMPS workflows.

npx -y skills add https://github.com/deepmodeling/deepmd-kit/tree/master/skills \
    --skill deepmd-install -y

If direct GitHub access fails, clone the official Gitee mirror and install from the local checkout:

git clone --depth 1 \
    https://gitee.com/deepmodeling/deepmd-kit.git \
    deepmd-kit-skill-source
npx -y skills add ./deepmd-kit-skill-source/skills \
    --skill deepmd-install -y

📚 Documentation and community

Citation

If DeePMD-kit contributes to published work, cite the general software paper that matches the version used and the method-specific papers listed in CITATIONS.bib:

  • Wang et al., “DeePMD-kit: A deep learning package for many-body potential energy representation and molecular dynamics,” Computer Physics Communications 228 (2018), 178–184 (describes the initial version). doi:10.1016/j.cpc.2018.03.016 Citations
  • Zeng et al., “DeePMD-kit v2: A software package for Deep Potential models,” The Journal of Chemical Physics 159 (2023), 054801 (covers features until v2.2.3). doi:10.1063/5.0155600 Citations
  • Zeng et al., “DeePMD-kit v3: A Multiple-Backend Framework for Machine Learning Potentials,” Journal of Chemical Theory and Computation 21 (2025), 4375–4385 (covers features until v3.0). doi:10.1021/acs.jctc.5c00340 Citations

License

DeePMD-kit is licensed under the GNU Lesser General Public License v3.0 or later.

Project links

Data verified by PyPI on Aug 19, 2026
Data provided by the project maintainers, verified at the time the release was uploaded to PyPI.

Key dates

PyPI data
Data sourced directly from PyPI's database.
  • Released:
Latest release

Owner

PyPI data
Data sourced directly from PyPI's database.

1 maintainer

PyPI data
Data sourced directly from PyPI's database.
Avatar for amcadmus from gravatar.com amcadmus

Credits

Author: DeepModeling

GitHub Statistics

Data verified by PyPI on Aug 19, 2026
The GitHub source repository was provided by the project maintainers and verified by PyPI at the time of upload. Stars, forks, and open issues/PRs are derived from that repository and have not been independently verified.

License expression

LGPL-3.0-or-later
View SPDX License List

Requires

Python >=3.10

Provides Extra

test dpa-adapt docs lmp ipi nvalchemi gui cu12 jax cpu gpu mpi torch

Tags

deepmd

Classifiers

Report project as malware

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

deepmd_kit-3.2.0.tar.gz (3.3 MB view details)

Uploaded Source

Built Distributions

If you're not sure about the file name format, learn more about wheel file names.

deepmd_kit-3.2.0-py37-none-win_amd64.whl (3.7 MB view details)

Uploaded Python 3.7Windows x86-64

deepmd_kit-3.2.0-py37-none-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (10.2 MB view details)

Uploaded Python 3.7manylinux: glibc 2.27+ x86-64manylinux: glibc 2.28+ x86-64

deepmd_kit-3.2.0-py37-none-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl (4.9 MB view details)

Uploaded Python 3.7manylinux: glibc 2.27+ ARM64manylinux: glibc 2.28+ ARM64

deepmd_kit-3.2.0-py37-none-macosx_11_0_x86_64.whl (4.4 MB view details)

Uploaded Python 3.7macOS 11.0+ x86-64

deepmd_kit-3.2.0-py37-none-macosx_11_0_arm64.whl (4.6 MB view details)

Uploaded Python 3.7macOS 11.0+ ARM64

File details

Details for the file deepmd_kit-3.2.0.tar.gz.

File metadata

  • Download URL: deepmd_kit-3.2.0.tar.gz
  • Upload date:
  • Size: 3.3 MB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for deepmd_kit-3.2.0.tar.gz
Algorithm Hash digest
SHA256 0a3b57d350c2fba26005e9696be1d598d2aea3a76f0fa947d3f63f96220f3530
MD5 190e88dee12669a97a1f52c60276af61
BLAKE2b-256 e11b9ec50c501e5b71ddbf6512c2d6f4463f89974a7f89e95fdb518fd5cbb82b

See more details on using hashes here.

Provenance

The following attestation bundles were made for deepmd_kit-3.2.0.tar.gz:

Publisher: build_wheel.yml on deepmodeling/deepmd-kit

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file deepmd_kit-3.2.0-py37-none-win_amd64.whl.

File metadata

  • Download URL: deepmd_kit-3.2.0-py37-none-win_amd64.whl
  • Upload date:
  • Size: 3.7 MB
  • Tags: Python 3.7, Windows x86-64
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for deepmd_kit-3.2.0-py37-none-win_amd64.whl
Algorithm Hash digest
SHA256 c660c234c88cd45a635f40cb5d86637102af7d2209ae599574d86579d52e5e6c
MD5 ab193a60e14957f885d9d0849f5c89e3
BLAKE2b-256 25b039b68a3c3ee5cbfbab1a7633244bc7ca3095c8874db7c32f3b24833f9bff

See more details on using hashes here.

Provenance

The following attestation bundles were made for deepmd_kit-3.2.0-py37-none-win_amd64.whl:

Publisher: build_wheel.yml on deepmodeling/deepmd-kit

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file deepmd_kit-3.2.0-py37-none-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl.

File metadata

File hashes

Hashes for deepmd_kit-3.2.0-py37-none-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl
Algorithm Hash digest
SHA256 331c45ffe85ecbaec1e3be03dae74be66ef8b9273c66d161e8c067d1cb23431c
MD5 0d3a1cc5bda4e40cd31fce70a9ac85f2
BLAKE2b-256 ca6933129fde3dfa8f95755e64f37b326587b68d9bec65d00f0a1a66db2dace2

See more details on using hashes here.

Provenance

The following attestation bundles were made for deepmd_kit-3.2.0-py37-none-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl:

Publisher: build_wheel.yml on deepmodeling/deepmd-kit

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file deepmd_kit-3.2.0-py37-none-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl.

File metadata

File hashes

Hashes for deepmd_kit-3.2.0-py37-none-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl
Algorithm Hash digest
SHA256 dd14bf4a4c32ea58048460b6317c326e1a1b01875c9d788f802564a6132aba25
MD5 a3f740c1e6e86955916a204ff28e67b2
BLAKE2b-256 9b8bba8c14dfebf7a9aac446065021ce9137a0f714217b052c0237d4c0d29e64

See more details on using hashes here.

Provenance

The following attestation bundles were made for deepmd_kit-3.2.0-py37-none-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl:

Publisher: build_wheel.yml on deepmodeling/deepmd-kit

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file deepmd_kit-3.2.0-py37-none-macosx_11_0_x86_64.whl.

File metadata

File hashes

Hashes for deepmd_kit-3.2.0-py37-none-macosx_11_0_x86_64.whl
Algorithm Hash digest
SHA256 7b3821a58f2d2e32786352bee6cbccf47fc1f59a25596a3880e53dbd6bb8b074
MD5 d7194d53e2d0aecc8e3025b31fc4cd8f
BLAKE2b-256 a80e9f2706a238e53c42ed1c85f01e369682e374ec6cd525574e892b930f9624

See more details on using hashes here.

Provenance

The following attestation bundles were made for deepmd_kit-3.2.0-py37-none-macosx_11_0_x86_64.whl:

Publisher: build_wheel.yml on deepmodeling/deepmd-kit

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file deepmd_kit-3.2.0-py37-none-macosx_11_0_arm64.whl.

File metadata

File hashes

Hashes for deepmd_kit-3.2.0-py37-none-macosx_11_0_arm64.whl
Algorithm Hash digest
SHA256 23200bdc8903a775209e021e54aa346ad201d7f75322b48c1b69a59a81f13d9c
MD5 5939c24fae1e5daad974442852453c78
BLAKE2b-256 918df10dc1fa30e45e949ff271e82d00bbdab7cd9e835f57b711beda234cf520

See more details on using hashes here.

Provenance

The following attestation bundles were made for deepmd_kit-3.2.0-py37-none-macosx_11_0_arm64.whl:

Publisher: build_wheel.yml on deepmodeling/deepmd-kit

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Release history Release notifications | RSS feed

This release

3.2.0 This release

6 files

Pre-release

3.2.0b0

6 files

3.1.3

6 files

3.1.2

6 files

3.1.1

6 files

3.1.0

6 files

Pre-release

3.1.0rc0

6 files

Pre-release

3.1.0a0

6 files

3.0.3

6 files

3.0.2

6 files

3.0.1

6 files

3.0.0

6 files

Pre-release

3.0.0rc0

6 files

Pre-release

3.0.0b4

6 files

Pre-release

3.0.0b3

6 files

Pre-release

3.0.0b2

6 files

Pre-release

3.0.0b1

6 files

Pre-release

3.0.0b0

6 files

Pre-release

3.0.0a0

6 files

2.2.11

6 files

2.2.10

6 files

2.2.9

6 files

2.2.8

6 files

2.2.7

6 files

2.2.6

6 files

2.2.5

6 files

2.2.4

6 files

2.2.3

6 files

2.2.2

5 files

2.2.1

5 files

2.2.0

5 files

Pre-release

2.2.0b0

5 files

2.1.5

6 files

2.1.4

6 files

2.1.3

6 files

2.1.2

1 file

2.1.1

6 files

2.1.0

6 files

2.0.3

5 files

2.0.2

5 files

2.0.1

5 files

2.0.0

5 files

Pre-release

2.0.0b4

5 files

Pre-release

2.0.0b3

4 files

Pre-release

2.0.0b2

4 files

Pre-release

2.0.0b1

3 files

Pre-release

2.0.0b0

4 files

Pre-release

2.0.0a1

Pre-release

2.0.0a0

4 files

1.3.3

4 files

1.3.2

4 files

1.3.1

4 files

1.3.0

4 files

1.2.4

4 files

1.2.3

4 files

1.2.2

4 files

1.2.1

3 files

1.2.0

3 files

1.1.5

4 files

1.1.4

3 files

1.1.3

3 files

1.1.2

3 files

1.1.1

4 files

1.1.0

4 files

1.0.2

4 files

1.0.1

4 files

1.0.0

4 files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page