Fair pricing for your ML metadata management

Some of your team are power users. Others only pop in from time to time. And your needs are constantly changing. So paying per user simply doesn’t make sense. It’s wasteful and inflexible at best. At worst, it’s plain unfair.

With Neptune, you get unlimited users on every paid plan. You only pay a flat monthly fee based on the duration of your experiments, model registry, or evaluation runs. We measure this time in ‘logging hours’.

Individual
$0/month
  • 1 user
  • 200 logging hours/month
Team
$150/month
  • Unlimited users
  • 1500 logging hours/month
  • Email support
  • Chat support
Organization
$600/month
  • Unlimited users
  • 6000 logging hours/month
  • Email support (1 business day)
  • Chat support
  • User access management
Custom
Custom
  • Unlimited users
  • Volume discounts (including unlimited usage)
  • User access management
  • SAML-based single sign-on (SSO)
  • Customizable contract and SLA
  • Dedicated support and onboarding
Organization
$600/month
(Billed annually)
  • Unlimited users
  • 6000 logging hours/month
  • Email support (1 business day)
  • Chat support
  • User access management
Custom
Custom
  • Unlimited users
  • Volume discounts (including unlimited usage)
  • Dedicated support and onboarding
  • Customizable contract and SLA
  • LDAP/SSO
“That sounds like a great deal.

But how do I guesstimate my usage?

If you train your models in the cloud, you can get a good idea by looking at your monthly usage. Cloud services like AWS charge you for the duration of your runs. In most cases, your Neptune logging hours will be the same.

If you’re still unsure, or you train models on your own infrastructure without proper monitoring, we’ll help you work out which plan is best for you with a quick analysis of your ML operations.

Your workflow is flexible. Your metadata management should be too.

avatar lazyload
quote
“An important detail that we considered when we decided to choose Neptune is that we can invite everybody on Neptune, even non-technical people like product managers — there is no limitation on the users. This is great because, on AWS, you’d need to get an additional AWS account, and for other experiment tracking tools, you may need to acquire a per-user license.”
Gianmario Spacagna Director of AI at Brainly
avatar lazyload
quote
Neptune made sense to us due to its pay-per-use or usage-based pricing. Now when we are doing active experiments then we can scale up and when we’re busy integrating all our models for a few months then we scale down again.
Viet Yen Nguyenh Senior Data Engineer at Zoined
avatar lazyload
quote
Pay-as-you-go is a practical pricing model for us because some months we will be running lots of pipelines and other months just common ones. So, when we can adjust the amount of computation or logging time, what we get from Neptune is a good solution.
Wojtek Rosiński Chief Technology Officer at ReSpo.Vision

No, you don’t pay for time spent looking at the UI. And other answers…

  • Yes! You can use Neptune for free for work, research, and personal projects. 

    On the Individual plan, you receive a monthly quota of 200 monitoring hours and a storage limit of 100 GB.

    If you are a student, professor, academic researcher, or Kaggle competitor, check out our Free Neptune Team program.

    And all our paid plans start with a 14-day free trial. No credit card needed

  • That’s correct. You can invite your entire ML team, your product division, CEO, and even your company dog. The price won’t change. The whole team really means the whole team.
  • We offer more than just a discount. If you’re a student, professor, Kaggle competitor, or researcher – you’ll get free access to our Neptune Team license with an increased number of monthly logging hours.
  • Absolutely. With the most competitive pricing model on the market, many startups chose Neptune over the competition. And don’t worry, you won’t get bumped up just because you raised Series A. We know what it’s like to be an early-stage SaaS. So we got you covered!
  • Yes! Unused hours for any given month get added to the following month’s quota for up to 12 months. After a year, unused logging hours expire.
  • You can easily top up your account by purchasing more hours. Each additional package of 100 logging hours costs $10. 

    Your account will enter read-only mode until you purchase additional logging hours. You’ll still be able to view, analyze, and fetch your tracked metadata using the read-only connection mode. But you won’t be able to track new runs or upload new metadata without buying more time.

  • If you have “hardware monitoring” enabled (the default), your logging hours equal the entire duration of your runs.

    If you choose to turn off “hardware monitoring”, you only pay for the amount of time your runs actually log metadata into Neptune’s database. In this case, your total logging hours equal the cumulative length of all the logging sessions over the course of your run. 

    Here’s how that works:

    If < 10 minutes pass between your 1st and 2nd log, the time between the 2 logs gets recorded as one continuous logging session, and gets added to your billable hours.

    Logging hours-8 mins

    Billable time: 8 mins

     

    And then if the 3rd logging event occurs within 10 minutes of the 2nd log, your logging session gets extended, and the time between the 2nd and 3rd logs gets added to your billable hours. And the same for the 4th log, and so on…

    Logging hours 18 mins

    Billable time: 18 mins

     

    But, if there’s more than a 10-minute gap before and after a logging event, only 1 second of logging time gets added to your bill for that log.

     

    Logging hours-2 seconds

    Billable time: 2 seconds

     

    So without hardware monitoring enabled, you only pay for the time you log metadata, not the entire run duration.

    Logging hours

     

  • Each workspace comes with 100 GB of storage space

  • You can easily top up your account by purchasing more storage space. Each additional package of 100GB costs $10. Alternatively, you can delete some old data to free up space within your current plan.  

    Your account will enter read-only mode until you purchase additional storage. You’ll still be able to view, analyze, and fetch your tracked metadata using the read-only connection mode. But you won’t be able to track new runs or upload new metadata without buying more storage space.

  • Absolutely not. Only the size of the metadata that you actually store in Neptune counts against your storage quota.
  • Neptune is currently hosted on the Google Cloud Platform Europe.
  • Yes! You can deploy Neptune on your own servers. Check out the “Your Server” tab above the pricing table at the top of the page. 
  • Whatever assistance you need, one of our friendly customer support team will be happy to help. That includes walking you through the setup, answering your product-related questions, debugging issues, or setting up an onboarding session for your new teammates.

Over 30,000 Data Scientists and ML Engineers use Neptune to effortlessly manage metadata

    Contact with us

    This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
    * - required fields