Sold by
Comet - Licensing only
Comet's machine learning platform integrates with your existing infrastructure and tools so you can reproduce, debug, manage, visualize, and optimize model - from training runs to production monitoring. Add two lines of code to your notebook or script and automatically start tracking code, hyperparameters, metrics, and more, so you can compare and reproduce training runs.
Reviews (36)
Consulting
Comet.ml Makes Experiment Tracking Effortless with Clear Dashboards
Reviewed on Sep 01, 2026
Review provided by G2
What do you like best about the product?
The best thing about Comet.ml is that it takes a lot of the manual effort out of experiment tracking. It's easy to see how different model runs performed, and the dashboards make it simple to spot trends without digging through logs. It has definitely helped make my workflow more organized. The automatic logging of metrics, parameters, and training runs has also made it much easier to compare experiments
What do you dislike about the product?
Although Comet integrates with major frameworks like PyTorch and TensorFlow, highly customized ML pipelines actually require additional logging and instrumentation work which is not good
What problems is the product solving and how is that benefiting you?
Since my work involves People Consulting, competency frameworks, working with large datasets of JDs/ competencies, using comet helps me in bringing more structure to this process as I can track experiments and their results in 1 place instead of relying on scattered files or notes or manually maintained records.
The biggest benefit is visibility and reduction of time spent manually.
The biggest benefit is visibility and reduction of time spent manually.
Sangeeta S.
Keeps ML experiments organized and comparable.
Reviewed on Aug 30, 2026
Review provided by G2
What do you like best about the product?
I really like that Comet.ml allows me to track experiment metrics, parameters, model versions, and training outputs all in one place. When I make changes to features or model parameters, comparing runs becomes easy-eliminating the need to take separate notes or maintain spreadsheets. The dashboard also lets me see at a glance which experiment is performing better.
What do you dislike about the product?
The platform is packed with features. Because of this, I initially found the user interface a bit cluttered. It can take some time to learn how to organize logging effectively, and setting things up for large projects might also require some time.
What problems is the product solving and how is that benefiting you?
Avoid losing track of your machine learning experiments with Comet.ml. We often test multiple models, datasets, and hyperparameters simultaneously. Comet.ml enables us to keep a record of what was changed and the resulting output, allowing for quick and efficient model comparisons. It is also excellent for collaboration, as others can see previous experiments without any extra effort.
Nilesh C.
Clear Visibility Into Experiments, Metrics, and Model Performance
Reviewed on Aug 29, 2026
Review provided by G2
What do you like best about the product?
It gives good visibility into experiments, metrics, and model performance without making the workflow complicated.
What do you dislike about the product?
The main thing I would improve is the learning curve for new users. Some features and settings can take a little time to understand, especially when managing a large number of experiments. The overall experience is good, but the initial setup and navigation could be more straightforward.
What problems is the product solving and how is that benefiting you?
Comet.ml helps solve the problem of keeping machine learning experiments organized and easy to compare. Instead of manually tracking different runs, metrics, and model versions, everything can be monitored in one place. This makes it easier to understand what is working, compare experiments, and avoid losing track of previous results. It saves time and makes the overall ML workflow more manageable.
Aggunuru V.
Powerful Experiment Tracking and ML Workflow Management
Reviewed on Aug 27, 2026
Review provided by G2
What do you like best about the product?
What I like most about Comet.ml is how easy it is to track experiments and how clearly it visualizes model performance. It makes it straightforward to compare runs, monitor metrics, keep results organized, and collaborate with team members. Overall, it helps make my machine learning workflow more efficient and reproducible.
What do you dislike about the product?
The main thing I dislike is that Comet.ml can have a bit of a learning curve for new users. Some of the more advanced features and dashboard configurations can feel overly complex, especially when you’re managing a large number of experiments. A simpler interface, along with more customization options, would make the overall experience better.
What problems is the product solving and how is that benefiting you?
Comet.ml helps solve the challenge of managing and tracking machine learning experiments, including parameters, metrics, datasets, and model versions. It lets me compare experiments more easily, reproduce successful results, and collaborate better with others. Overall, it saves time, reduces the need for manual tracking, and makes the ML development process more organized and efficient.
Arun R.
Comet.ml Makes ML Experiment Tracking and Visualization Effortless
Reviewed on Aug 26, 2026
Review provided by G2
What do you like best about the product?
What I like most about Comet is that it keeps the entire ML experimentation process organized in one place. It makes it easy to track experiments, compare different runs, and see how changes in parameters or code affect model performance. I also like the visualization and model versioning capabilities because they make it much easier to understand results and reproduce successful experiments. For LLM and AI projects, the Opik capabilities are also useful for tracing and evaluating model and agent behavior.
What do you dislike about the product?
The main downside for me is that Comet has quite a lot of features, so it can take some time to understand the platform and decide which features are actually needed for a particular project. The interface can also feel a little overwhelming when managing a large number of experiments, metrics, and artifacts. A simpler onboarding experience and more streamlined navigation would make it easier for new users to get started.
What problems is the product solving and how is that benefiting you?
Comet helps solve the problem of managing and reproducing machine learning experiments as projects become more complex. Instead of manually keeping track of parameters, metrics, code versions, datasets, and models, everything can be linked and tracked in one place. This makes it much easier to compare experiments, identify what worked, reproduce results, and collaborate with other team members. It also provides useful observability and evaluation capabilities for LLM and AI applications through Opik
Vignesh A.
Great for Beginner Developers Learning ML & LLMs, Though Some Gaps Remain
Reviewed on Aug 25, 2026
Review provided by G2
What do you like best about the product?
The best way for a beginner developer to understand ML and LLMs.
What do you dislike about the product?
Being more transparent leads to security threats.
What problems is the product solving and how is that benefiting you?
Supports entire development project using ML
Mohammed Mudasser
Experiment tracking has improved collaboration and simplifies comparing model iterations
Reviewed on Aug 18, 2026
Review from a verified AWS customer
What is our primary use case?
My main use case for Comet is ML experiment tracking and evaluation. A specific example of how I use Comet is that I typically track model runs, compare the metrics and the hyperparameters, and keep the experiments organized so I can easily see what worked and reproduce the better results.
What is most valuable?
What I find particularly useful about my main use case with Comet is having the experiment history, metrics, parameters, and the other runs details organized in one place.
The best features Comet offers in my opinion are the visual dashboards, the run comparison, and the experiment tracking in ML, and those features stand out for me because they automatically capture things such as parameters, code, metrics, and system information, making it much easier to understand and reproduce experiments.
Comet has positively impacted my organization by mainly improving collaboration and productivity by keeping the experiment results, the metrics, and the model comparisons in one place, making it much easier to share results with others and quickly identify which approach is performing better so we can iterate without spending as much time organizing experiments.
What needs improvement?
One area I would like to see Comet improve is in the ease of navigating and comparing a large number of experiments, as better filtering and quicker ways to find specific runs would make the workflow even smoother and save some manual effort.
The documentation part of Comet is generally very useful, but some advanced features could be explained more simply for new users, and I would also appreciate even smoother navigation and comparison when working with a large number of experiments.
I chose a rating of 8 out of 10 because Comet covers the core experiment tracking needs really well, especially tracking and comparing runs, and to make it a 10, I would mainly want smoother navigation for bigger projects and more intuitive guidance around some of the advanced features.
For how long have I used the solution?
I have been using Comet for around a year.
What do I think about the stability of the solution?
Comet's tracking and evaluation results are reliable and consistent with what I expect from my experiments, and having the metrics and run details recorded clearly also makes it easier to validate results rather than relying on assumptions.
Comet is stable and reliable during my usage, especially for logging metrics and tracking the experiments across different runs.
What do I think about the scalability of the solution?
I found Comet to be quite scalable for the projects I have worked on, including handling larger workloads without major issues, and as the number of experiments grown, the navigation can take a little more effort, but overall, it has handled my use case reliably.
Which solution did I use previously and why did I switch?
I did not previously use a different solution before Comet.
What was our ROI?
I have seen a positive return on investment with Comet in my experiment management process, as it has definitely reduced the manual work and helped me iterate on models faster.
Which other solutions did I evaluate?
Before choosing Comet, I evaluated a few alternatives, mainly MLflow, and we ended up preferring Comet because its experiment comparison, visualization, and overall workflow felt more convenient for our use case.
What other advice do I have?
I appreciate the integrations with Comet because it fits into the ML tools I am already using rather than forcing a completely different workflow, and the collaboration features are also useful for sharing experiments and comparing results with teammates, especially when multiple people are working on model iterations.
I primarily use Comet as a cloud-based platform, accessed through its web interface and integrated into my ML workflows, and I am not directly involved in managing the infrastructure and the deployment configuration part.
I mainly use AWS alongside Comet for my machine learning workflows and cloud infrastructure, with Comet itself serving as the experiment tracking layer while the compute and other supporting services run under Amazon Web Services.
I would recommend Comet to teams that want a simple way to track, compare, and organize machine learning experiments, and if you are running multiple model iterations, I would definitely try it because the experiment history and the visualization make the workflow much easier to manage.
Rakesh G.
Comet Makes ML Experiment Tracking and Team Collaboration Effortless
Reviewed on Aug 12, 2026
Review provided by G2
What do you like best about the product?
What I like most about Comet is how it helps me organize, compare, and reproduce ML experiments. It also has a clear dashboard and collaboration features, which make it much easier for the team to understand and stay aligned.
What do you dislike about the product?
What I dislike is the complexity involved in tracking and setting up an experiment during the initial stage.
What problems is the product solving and how is that benefiting you?
Comet solves the problem of managing ML experiments, metrics, parameters, and model versions all in one place. This makes it easier to compare results, and it helps me identify which approach is performing best.
Oil & Energy
Centralized ML Experiment Tracking with Clear Dashboards and Strong Reproducibility
Reviewed on Aug 10, 2026
Review provided by G2
What do you like best about the product?
This platform makes it easy to track and compare machine learning experiments all into a centralised workspace. I also like how it automatically records the needed parameters, metrics, code changes and other system information without even requiring any extensive support. Their dashboard provide clear visual comparison between the different model runs. It also helped our model registry and artefact versioning, improving the reproducibility and team collaboration. Overall, I would say this gave us better control over the complete model development life cycle.
What do you dislike about the product?
Their interface can be bit overwhelming initially because of the number of features and configuration options available. And for large projects with multiple experiment may also require careful organization to keep their workspace manageable. Some advanced capabilities and other team management features are limited to higher priced plans and uploading extensive logs and artifacts can also add more storage and overhead the performance. Having a better onboarding and simpler cost visibility or transparency would improve the overall platforms experience.
What problems is the product solving and how is that benefiting you?
This platform solves the difficulty of manually tracking model experiments across their spreadsheet and even, notebooks and disconnected tools. It gave us a reliable record of every model run and including its parameters, metrics code and other related artefacts. This allowed our team to reproduce successful experiments and understand why one model performs better than the other one. Their model registry also create a more structured process for promoting models into production, as a result of it our development became faster and more collaborative and less prone to repetitive work.
Anil B.
Simple, All-in-One Machine Learning Experiment Tracking with Comet.
Reviewed on Aug 09, 2026
Review provided by G2
What do you like best about the product?
The main thing that I like in Comet.ml is that it is really simple to track the results of my machine learning experiments. With Comet.ml, I can compare different models, track metrics, save the parameters, and organize the results of the experiments all in one place. The dashboards are useful for evaluating the performance of the models.
What do you dislike about the product?
The reason why I don’t like the platform of Comet.ml is that there are some complex features which need some time to grasp. Besides, the interface might be quite complicated when dealing with several experiments and the need to configure some functions. I believe that the customization options of the reporting and dashboard can be more versatile.
What problems is the product solving and how is that benefiting you?
Comet.ml addresses the issue of having to do all of the experiment management and comparisons manually. It stores all the experiment parameters, metrics, models, and results together in one place. This allows me to keep track of my progress, figure out which models work better, and replicate experiments more efficiently.