NVIDIA Deep Learning AMI Reviews & Product Details

The best part of having AMI is we can switch up the GPU accelerations on VM in a couple of minutes when required and according to the load on the servers. It is also very helpful in managing the cost of development during training of deep learning models, and we can switch to any Nvidia GPU on the go as and when required. During product prototyping, the available SDKs, pre-trained models and other applications are very helpful in delivering the first version to the clients in less time and without deep research. Review collected by and hosted on G2.com.

Nvidia can improve on the UX/UI side, and we are waiting for more pre-trained models for the robotics arm and some ready models for AR/VR and animations with characters. Machines are so powerful and efficient that NVIDIA has no space for disappointment. Review collected by and hosted on G2.com.

The performance is superb. Access is fast and easy. NVIDIA accelerates innovation by eliminating the complex task of building and optimizing a complete deep-learning software stack tuned specifically for GPUs.It is a great platform for learning deep-learning neural networks Review collected by and hosted on G2.com.

It can be used for classification problems. Lack of common sense, emotions are not handled properly now. It can be improved to handle more data. Some of the standard deep learning algorithms are difficult to execute using this software Review collected by and hosted on G2.com.

Its having more exposure. next generation of learning. AI, Data Science & graphics learning which is making it more powerful.

Nvidia the name for graphics now is going to bring revolution to the world. Review collected by and hosted on G2.com.

nothing to dislike anything as such. just waiting for more amazing learning in a conceptual way. I know Nvidia is not going to disappoint in this. Lets wait and watch Review collected by and hosted on G2.com.

It runs diverse deep learning methods like data science and containers and can be adjusted as per your requirement. Review collected by and hosted on G2.com.

needed a more active community to resolve the issues in no time. Review collected by and hosted on G2.com.

This is one of the best application for the Analysis of Deep Learning. There are many model which can be tried and provide a good optimal result Review collected by and hosted on G2.com.

There is noting as such but the UI and model complexity can be reduced a bit so that it can become a light application. Review collected by and hosted on G2.com.

Easy to use with abundant documentation. The support community was also very useful in solving issues. Review collected by and hosted on G2.com.

The price is a little too high for our liking. The product features are also a little too limited. Review collected by and hosted on G2.com.

I like the best that they have a very well-organized environment that helps in running various deep learning processes like data science, and containers, and can be tuned as per your requirement in the development. Review collected by and hosted on G2.com.

Real-time support can be improved & made more effective for the consumers Review collected by and hosted on G2.com.

AMI works on the principles of cloud technology which are highly secured and zero chances of data corruption. NVIDIA deep learning has unique programmed software which utilizes all the tensor cores making PCs much faster than the conventional ones. Review collected by and hosted on G2.com.

As a foreigner user of NVIDIA AMI, 3$ per hour is too much which is the only downside of this technology but could be warehouses available here so rates should be according to country's currency. Review collected by and hosted on G2.com.

Incredible performance and models for deep learning. Review collected by and hosted on G2.com.

UX complexity has proven to be a challenge with my team. Review collected by and hosted on G2.com.

This container comes pre-optimized, no need to set up a virtual env in your on-site machine (though sometimes that is definitely better). You also get access to the top-grade Nvidia GPUs (non-retail). On standard Ubuntu, and you can obviously load up the GUI if you want, but otherwise, very familiar OS which has minimal (none?) collision issues with most any recent AI/ML libraries (unlike Windows say) Review collected by and hosted on G2.com.

The difference between building your own box and spinning one up in a cloud marketplace becomes felt the longer you work on it and exactly how much time you will need to run it every day. I find that having a box local can be helpful for experimentation, but otherwise it'll cost you 3-4 USD/hr to run this instance. The highest grade GPU (retail) is currently the RTX 3090, which costs MSRP < $2000. If you already have a nice machine and you wanted to just purchase the card, it may make more sense to do that (though currently we are in a chip shortage so maybe not exactly at this moment). Still, a process that may require multiple boxes, unless you want to setup a server farm in your home, this is the best way to go. Review collected by and hosted on G2.com.