52 neptune.ai Reviews
Overall Review Sentiment for neptune.ai
Log in to view review sentiment.
Neptune AI covers all the needs of a Model Registry in an efficient way, thus, it can be a core part of the lifecycle of ML models within a platform.
It offers a user-friendly UI with which, in addition to being able to easily view the registered metadata associated with each execution, we can build more or less complex dashboards. On the user side, the API works in an easy, intuitive and flexible way, providing Data Scientists a simple way to interact with the projects from both script and Jupyters Noteboks.
About the installation, it is as simple as installing a Python package if you use the SaaS option. If you opt for something self-hosted it is a bit more complex but you will always have expert support.
Finally, the Neptune team always responds quickly, offering both mail support and call availability, although most of the time the doubts can be solved by consulting the documentation which is more than correct. Review collected by and hosted on G2.com.
As I have mentioned in the benefits, the use of the Neptune API is very flexible, perhaps excessively flexible if used in a corporate environment since we could find metadata structured in different ways in different projects.
If this is a problem, as a company you may have to develop custom components on top of the Neptune AI API to adapt it to your specific needs and standardize the metadata registry. Review collected by and hosted on G2.com.
- Excellent customer support
- Good documentation
- Easy to use API and implement
- Great UI
- Good service availabilty (almost no downtime)
- Easy to integrate with different tools Review collected by and hosted on G2.com.
- Limited functionality for creating reports
- No integrated model registry Review collected by and hosted on G2.com.
The platform is straightforward to use and seamlessly integrates into your code, with helpful tutorials available for guidance. Comparing experiments and model metrics is incredibly easy, and the use of tags makes refining your analysis more efficient. Review collected by and hosted on G2.com.
The dashboards are useful, but seamless integration with third-party platforms like Tableau, which we frequently use for reporting, would be a valuable addition. Review collected by and hosted on G2.com.
What stands out most about Neptune.ai is its unparalleled ease of use and versatility as a real-time database, perfectly catering to ML projects and a variety of other data-intensive tasks. Its interface is exceptionally intuitive, enabling effortless exploration, comparison, and analysis of metrics across numerous runs. This feature alone significantly enhances our workflow, allowing us to swiftly identify trends, make informed decisions, and fine-tune our models. Furthermore, Neptune's capability to manage simultaneous logging from thousands of runs seamlessly has been pivotal in scaling our operations efficiently. The platform has rapidly become the cornerstone of our workflow, serving not just as a logging tool but also as the primary repository for our metadata and artifacts, ensuring consistency and accessibility across our projects. Review collected by and hosted on G2.com.
My only minor gripe with Neptune.ai is its limited native capabilities for crafting complex plots, especially when it comes to combining and visualizing metrics together, which would enhance our ability to analyze data deeply. Review collected by and hosted on G2.com.
We've been using Neptune for two years now (we have self-hosted version). I use it almost on daily basis. During this time, it has introduced many new features and found by my team bugs have been fixed. Neptune integrates easily into our workflow and offers broad compatibility with many popular frameworks like PyTorch or AirFlow. This makes it easy to integrate it inside your code and quicker to start using.
For our company, a big plus of Neptune is the availability of a self-hosted version. We were looking for such a solution and found that not many services offer this option at an affordable price. Using the self-hosted version of Neptune is no different from the cloud version for end users. However, it has a several of advantages - unlimited storage and isolation from the global network, which increases data security.
In the end, Neptune has made our workflow more reproduceable, straightforward, visual test. Review collected by and hosted on G2.com.
At the begning of our usage, this tool had bugs, UI wasn't be reach, but the Neptune support/development team fixed these issues, enchanced UI. Currently, we have no significant complaints about Neptune.
Maybe the only downside is that new features are delivered somewhat slower for self-hosted version compared to the cloud version, but they are trying to speed up the delivery of new features to us. Review collected by and hosted on G2.com.
Neptune.ai has been a lifesaver for our AI research team when it comes to ensuring reproducibility. When exploring new research or working on project fine-tuning tasks, it's crucial for us to be able to reproduce previous results accurately. With Neptune.ai, we're able to store all of our experimental configurations, so we can easily re-procedure our models and collaborate with other researchers on an intuitive UI.
Besides, what has really impressed us is Neptune.ai's excellent support team. Whenever we have faced any issues, they have been quick to respond and are always willing to go above and beyond to help us resolve our problems. We're grateful to have found Neptune.ai, and we highly recommend it to anyone who is looking for a tool that can enhance their model reproducibility and teamwork. Review collected by and hosted on G2.com.
- Could make a better Model Registry, e.g. add markdown documentation for each model, add discussions, ... Review collected by and hosted on G2.com.
I like the project-based structure which helps to keep the overview and to collaborate. It's easy to integrate neptune into your ML framework and new features are released in a short update cycle. The customer support is remarkable. Review collected by and hosted on G2.com.
Because of the short update cycles code is easily getting old. Some neural network specific features are still to be implemented like weight histograms, network graphs, and visualizations of embeddings. Review collected by and hosted on G2.com.
It's a nice tool focused on what it does best: experiment and artifacts tracking. No boilerplate, very focused, very lightweight. Contrary to some other competitors, it's not bloated with half-backed features and it's not trying to conquer the whole MLOps sector. It is very easy to use and to integrate in any python-based experiment. I have been using it daily for a few years and I like it very much. Implementing it in my team was very straightforward and my team was able to start using it right away. Review collected by and hosted on G2.com.
The lineage feature is a bit primitive at the moment. Artifacts are linked to their experiments, but there is no artifact-centered view at the project level. Review collected by and hosted on G2.com.
The scalability of the product. The much improved interface and having tons of relevant features. Also the close support from their team. Review collected by and hosted on G2.com.
Some features are distracting and unnecessary depending on your usecase. Migrations to newer versions could have been simpler. Review collected by and hosted on G2.com.
