Compare Neptune vs Comet

Super similar software
Unsimilar price

vs

Both Neptune and Comet allow you to track, visualize, and compare your ML model metadata. Both have flexible deployment options. And both make easy collaboration a reality.

But only Neptune will fit comfortably into your budget.

Show me pricing Show me feature-by-feature comparison

Usage-based pricing plans
that evolve with your needs

Flexible data structures
for seamless migration

Clear, detailed docs
& tutorials

Faster ML metadata management. For less

Cost

Reasonable pricing that fits your reality

Let your team test Neptune’s functionality in a live project — with zero budget spend — on our free plan. Including 5 users and 1 active project.

As you scale, our usage-based paid plans evolve with your growing requirements. You can add additional active projects to your package as and when you need.

Interface

The user interface that works as fast as you do

You shouldn’t have to live with a lagging UI or long experiment load times. Neptune is — and always will be — everything you need to manage your ML metadata. And nothing else. We built it as lightweight as possible. So you can work as fast as possible.

Scale from 10 to 10000+ runs. With zero effect on speed.

See pricing Find out more about Neptune

Feature-by-feature comparison

Take a deep dive into
what makes Neptune different

Show differences only

Neptune

Comet

Commercial Requirements

Commercial requirements

Standalone component or a part of a broader ML platform?

Standalone component

Standalone component

Is the product available on-premises and / or in your private/public cloud?

Tracking is hosted on a managed server, and can also be deployed on-premises and in a public/private cloud server

Yes, you can deploy Comet on any cloud environment or on-premise

Is the product delivered as commercial software, open-source software, or a managed cloud service?

Managed cloud service

Managed cloud service

What is the pricing model?

Free individual and paid team/enterprise plans available

Free individual and paid team/teams pro/enterprise plans available

SLOs / SLAs: Does the vendor provide guarantees around service levels?

Email and chat support for individuals and teams. Well-defined support SLAs for enterprise customers

Email and Slack support for teams. Support SLAs for enterprise customers

Support: Does the vendor provide 24×7 support?

Only for enterprise customers

Only for enterprise customers

SSO, ACL: does the vendor provide user access management?

Only for enterprise customers

Only for enterprise customers

Security policy and compliance

Yes

General Capabilities

Setup

What are the infrastructure requirements?

No special requirements other than having the neptune-client installed and access to the internet if using managed hosting. Check here for infrastructure requirements for on-prem deployment.

No special requirements other than having the comet_ml installed and access to the internet if using managed hosting. Check here for on-prem deployment

How much do you have to change in your training process?

Minimal. Just a few lines of code needed for tracking. Read more

Minimal. Few lines of code needed for tracking

Does it integrate with the training process via CLI/YAML/Client library?

Yes, through the neptune-client library

Yes, through comet_ml library

Does it come with a web UI or is it console-based?

Serverless UI

No

No

Flexibility, speed, and accessibility

Customizable metadata structure

How can you access model metadata?

– gRPC API

No

No

– CLI / custom API

– REST API

No

– Python SDK

– R SDK

No

– Java SDK

No

– Julia SDK

No

No

Supported operations

– Search

– Update

– Delete

– Download

Distributed training support

Pipelining support

Logging modes

– Offline

– Disabled/off

No

– Asynchronous

Yes by default

– Synchronous

No explicit selection

Live monitoring

Mobile support

No

No

Webhooks and notifications

No

Webhooks only for model management. Notifications only for change in experiment status

Experiment Tracking

Log and display of metadata

Dataset

– location (path/s3)

– hash (md5)

– Preview table

– Preview image

– Preview text

– Preview rich media

– Multifile support

Code versions

– Git

– Source

– Notebooks

Parameters

Metrics and losses

– Single values

– Series values

– Series aggregates (min/max/avg/var/last)

Tags

Descriptions/comments

Rich format

– Images (support for labels and descriptions)

No

– Plots

– Interactive visualizations (widgets and plugins)

– Video

– Audio

– Neural Network Histograms

No

– Prediction visualization (tabular)

No

– Prediction visualization (image)

No

NA

– Prediction visualization (image – interactive confusion matrix for image classification)

No

– Prediction visualization (image – overlayed prediction masks for image segmentation)

No

– Prediction visualization (image – overlayed prediction bounding boxes for object detection)

No

Hardware consumption

– CPU

– GPU

– TPU

No

No

– Memory

System information

– Console logs (Stderr, Stdout)

– Error stack trace

No

– Execution command

No

No

– System details (host, user, hardware specs)

Environment config

– pip requirements.txt

– conda env.yml

– Docker Dockerfile

Files

– Model binaries

– CSV

– Any file

All files can be logged, but some formats might not be rendered

– External file reference (s3 buckets)

Comparing experiments

Table format diff

Overlayed learning curves

Parameters and metrics

– Groupby on experiment values (parameters)

– Parallel coordinates plots

– Parameter Importance plot

No

No

Rich format (side by side)

– Image

– Video

No

No

– Audio

No

No

– Plots

No

– Interactive visualization (HTML)

No

No

– Text

Yes

– Neural Network Histograms

No

No

– Prediction visualization (tabular)

– Prediction visualization (image, video, audio)

No

No

Code

– Git

No

No

– Source files

No

No

– Notebooks

No

Environment

– pip requirements.txt

No

No

– conda env.yml

No

No

– Docker Dockerfile

No

No

Hardware

– CPU

No

– GPU

No

– Memory

No

System information

– Console logs (Stderr, Stdout)

– Error stack trace

– Execution command

No

No

– System details (host, owner)

Yes

Data versions

– Location

– Hash

– Dataset diff

– External reference version diff (s3)

No

Files

– Models

No

No

– CSV

No

Custom compare dashboards

– Combining multiple metadata types (image, learning curve, hardware)

– Logging custom comparisons from notebooks/code

No

No

– Compare/diff of multiple (3+) experiments/runs

Organizing and searching experiments and metadata

Experiment table customization

– Adding/removing columns

– Renaming columns in the UI

Yes

– Adding colors to columns

Yes

– Displaying aggregate (min/max/avg/var/last) for series like training metrics in a table

No

– Automagical column suggestion

No

Experiment filtering and searching

– Searching on multiple criteria

– Query language vs fixed selectors

Query language

– Saving filters and search history

Custom dashboards for a single experiment

– Can combine different metadata types in one view

– Saving experiment table views

– Logging project-level metadata

– Custom widgets and plugins

No

Tagging and searching on tags

Nested metadata structure support in the UI

No

Reproducibility and traceability

One-command experiment re-run

No

Experiment lineage

– List of datasets used downstream

No

– List of other artifacts (models) used downstream

No

Limited to datasets

– Downstream artifact dependency graph

No

Limited to datasets

Reproducibility protocol

Is environment versioned and reproducible

Saving/fetching/caching datasets for experiments

No

No

Collaboration and knowledge sharing

User groups and ACL

Only in paid plans

Only for Teams Pro and Enterprise

Sharing UI links with project members

Sharing UI links with external people

Commenting

Interactive project-level reports

No

Model Registry

Model versioning

Code versions (used for training)

Environment versions

Parameters

Dataset versions

Results (metrics, visualizations)

Explanations (SHAP, DALEX)

Model files (packaged models, model weights, pointers to artifact storage)

Model lineage and evaluation history

Models/experiments created downstream

No

No

History of evaluation/testing runs

No

Support for continuous testing

No

No

Users who created a model or downstream experiments

No

No

Access control, model review, and promoting models

Main stage transition tags (develop, stage, production)

Custom stage tags

No

Locking model version and downstream runs, experiments, and artifacts

No

No

Adding annotations/comments and approvals from the UI

Model compare (current vs challenger etc)

No

Compatibility audit (input/output schema)

No

No

Compliance audit (datasets used, creation process approvals, results/explanations approvals)

No

No

CI/CD/CT compatibility

Webhooks

No

Model accessibility

No

No

Support for continuous testing

No

No

Integrations with CI/CD tools

No

Model searching

Registered models

Active models

By metadata/artifacts used to create it

By date

Yes

By user/owner

Yes

By production stage

Yes

Search query language

RegEx

Model packaging

Native packaging system

No

No

Compatibility with packaging protocols (ONNX, etc)

No

No

One model one file or flexible structure

No

No

Integrations with packaging frameworks

No

No

Integrations and Support

Languages

Java

No

Julia

No

No

Python

R

No

REST API

No

Model training

Catalyst

No

CatBoost

No

fastai

FBProphet

Gluon

No

No

HuggingFace

H2O

No

No

LightGBM

Paddle

No

No

PyTorch

PyTorch Ignite

No

PyTorch Lightning

Scikit Learn

Skorch

No

Spacy

No

Spark MLlib

No

Only with Spark NLP

Statsmodel

No

No

TesorFlow / Keras

XGBoost

Hyperparameter Optimization

Hyperopt

No

No

Keras Tuner

No

Optuna

No

Ray Tune

No

Scikit-Optimize

No

Model visualization and debugging

DALEX

No

Netron

No

No

SHAP

No

TensorBoard

IDEs and Notebooks

JupyterLab and Jupyter Notebook

Google Colab

Deepnote

AWS SageMaker

No

Data versioning

DVC

No

Orchestration and pipelining

Airflow

No

Argo

No

No

Kedro

No

Kubeflow

No

No

ZenML

Experiment tracking tools

MLflow

Sacred

No

TensorBoard

CI/CD

GitHub Actions

No

Gitlab CI

No

No

CircleCI

No

No

Travis

No

No

Jenkins

No

No

Model serving

Seldon

No

No

Databricks

No

No

Model versioning

Seldon

No

No

Fiddler.ai

No

No

Arthur.ai

No

No

LLMs

LangChain

No

This table was updated on 11 August 2023. Some information may be outdated.
Report outdated information here.

Neptune is way better than the other tools I’ve tried, like Comet and WandB. In my opinion, Neptune has the cleanest and most intuitive interface — that’s the main reason I prefer using it.

Klaus-Michael Lux Data Scientist

You deserve fairly priced ML metadata management

With Neptune, that’s exactly what you get.

Check out the best-fit plan for your business today.