Runpod Hybrid Cloud

What do you like best about the product?

Remote access to any hardware, even with the current market calamites.

What do you dislike about the product?

Sometimes hardware gets rented out too quickly while testing and restarting a pod, a small 1 minute grace period would be nice.

What problems is the product solving and how is that benefiting you?

Budget and Setup are solved.

What do you like best about the product?

I use RunPod to run models and tasks that I can't handle with the resources I have locally. I like the price and how practical, fast, and easy it is to turn pods on and off; plus, I find it user-friendly. The global environment gives me more alternatives to launch pods without traditional persistent storage. I also value the variety of GPUs to choose from and that there are cheaper GPU pods that work for me. The initial setup was easy, and the prompts they provide for the agent helped me.

What do you dislike about the product?

Sometimes I need 4 GPUs of a model for parallel processes, but it's almost impossible to get even 2. Also, I have to edit the SSH connection every time I start a new pod because the IP and port vary from one pod to another.

What problems is the product solving and how is that benefiting you?

I use RunPod to run models or tasks that I can't execute or load with the infrastructure I have locally.

What do you like best about the product?

The GPU availability and the pod architecture with Linux as a backbone.

What do you dislike about the product?

Low availability of certain GPUs sometimes.

What problems is the product solving and how is that benefiting you?

It allows me to test infrastructure without having to buy and run the hardware.

What do you like best about the product?

Deployment speed and pricing is the golden key; also, it's easy to use.

What do you dislike about the product?

Honestly, I love pretty much everything about RunPod; to be fair, you are the cheapest, fastest, most robust, and most secure—so I love it all :)

What problems is the product solving and how is that benefiting you?

As a Software Architect and the founder of Lunalux Bilişim, our core focus is research and development. RunPod plays a crucial role in our workflow by giving us the agility to rapidly prototype, build emerging technologies, and test new AI models—essentially allowing us to experiment and innovate with the curiosity and freedom of kids in a playground.

What do you like best about the product?

the deployment was really fast. the machines are stable and extremely fast. a variety of tools to run your commands.

What do you dislike about the product?

it could be cheaper for academic purposes

What problems is the product solving and how is that benefiting you?

ı run my computer vision tasks on the machines and the results are quite fast and accurate.

What do you like best about the product?

Easy to setup, and skills for agents does help me for something automation task (training, fine-tuning, and evaluating the models; most of those tasks are agents do it and report back to me)

I do not know if RunPod expose a webhook for notification, I would let my agent notify me via it too (you guys can cost it but please be reasonable).

What do you dislike about the product?

Not much to add right now.

One gripe, though: low-tier GPUs like the RTX 2000 Ada or RTX A4000 are often sold out. Sometimes all I need is a small GPU to test inference, but with nothing available, Colab ends up being my only option.

What problems is the product solving and how is that benefiting you?

It's easy to set up and use, which saves me time, and the price is reasonable.

What do you like best about the product?

Access to large quantities of compute power for big projects readily available

What do you dislike about the product?

The fact that GPU's can just fall off at any point in time because someone no longer runs it at that point in time of the day

What problems is the product solving and how is that benefiting you?

Large compute power at a time when the industry has a shortage and a high price point on compute and processing.

What do you like best about the product?

the pricing is precise, allowing you to pay only for what you consume.

What do you dislike about the product?

The initial setup is a bit complex but not insurmountable.

What problems is the product solving and how is that benefiting you?

I can run generation models or do training with custom-made stacks without having to buy a GPU or rent a server for a month.

What do you like best about the product?

I appreciate that options are all presented clearly and there arent "gotcha" moments. you are advised when choices need to made that they have downstream consequences so you never feel like you hit an unexpected brick wall. As an media designer it is very help to me that you do not need a computer science degree to navigate. I am collaborating with an AI on a project and between the guides and the AI ive been fortunate to hit no dead ends. Thankyou Runpod :)

What do you dislike about the product?

I honestly haven't encountered anything. Take my input with a grain of salt, I dont think I am in the core demographic/occupation for the intended user so my ability to comment on specifics is limited.

What problems is the product solving and how is that benefiting you?

Accessibilty to powerful AI model use. It is saving massive capital investment to even get my foot into some doors. This service has increased my capabilities tenfold.

What do you like best about the product?

I like the ease of use and the simple, self-descriptive UI of RunPod. I don't have to read a lot of documentation, which makes it user-friendly. The MCP server is easy to install and informative. My agents find it straightforward to set up without much troubleshooting. I also appreciate the serverless component, which allows us to scale usage throughout the day without constantly paying for infrastructure. Setting up was very easy, as I just copied the MCP config into Claude, and it was ready to log in and initialize effortlessly.

What do you dislike about the product?

I often struggle with getting capacity for my experiments, especially with the 50 90s, which seems to be due to a general lack of GPUs. There was an instance where I ran an experiment that produced results, but I couldn't restart the machine to retrieve those results because it was no longer available, and I had to abandon it. That situation was really not ideal.

What problems is the product solving and how is that benefiting you?

RunPod provides elastic GPU compute availability without the high cost of hyperscalers. It has an easy-to-use UI and MCP, allowing quick setup and serverless scaling based on daily usage, reducing infrastructure costs.