1. What each one is good at
| Your $20 plan | A rented GPU server | |
|---|---|---|
| Best at | Planning, explaining, reviewing, writing the first version, steering a project | Running a very large open model, running many helpers at once, long jobs, training |
| How you pay | A flat monthly price with a usage allowance (limits vary by company and plan) | By the second while it runs, plus storage. You stop paying when you terminate it |
| Limits | The allowance can run out on a big project | You manage the setup. Left running, it keeps billing |
| Privacy | Your prompts go to the plan's company | The model runs on a machine you rented. If privacy matters, check RunPod's security terms for the cloud type you pick |
2. Three ways to use the muscle
A. A big open model as your bulk coder
Rent one GPU, run an open model on it, and point your coding tools at it. Many coding tools let you set a custom OpenAI-compatible address; check that yours does. RunPod's documentation describes serving language models that way [1]. Your plan then handles the planning and review while the rented model does the typing.
B. A team of helpers on one rented machine
A server can run many helper programs at once, and one big model can answer all of them. Serving software such as vLLM is built for many requests at a time [1]. How many helpers one machine can keep busy depends on the model and the work, and we have not measured it. Helpers that call a paid hosted API still bill per use; the savings only appear if they call a model you are running yourself.
C. A burst for one big job
Rent a large machine for the hours a heavy task needs: a long training run, a huge batch of generated assets, or a model that will not fit anywhere at home. Terminate it when the job is done.
Connect your assistant to real data
An assistant gets much more useful when it can read real data instead of only talking. Many assistants can connect to outside sources through connectors: market data*, sports stats, weather, business numbers, your own spreadsheets. The assistant pulls the data, sorts it, compares it and explains what it sees in plain words. A rented server can chew through much bigger piles of data than a laptop can.
- A hobbyist asks which numbers actually changed since last season.
- A maker asks it to pull part prices from several shops and line them up.
- Someone who follows the markets asks for a plain summary of what moved today.*
What you do with the answers is up to you. Check anything that matters yourself.
* For education and information only. Not financial advice. We are not financial advisors. Trading and investing involve risk, including loss of principal.
3. What fits, and what it costs per hour
Memory figures are estimates from our VRAM calculator (weights, context memory, 10% working memory and 1.5 GB overhead per GPU). They are not measurements. Rates are RunPod on-demand Secure Cloud rates fetched within the last 30 days and can change.
| What you want to run | Estimated memory | A GPU setup that holds it | Per hour |
|---|---|---|---|
| A 32B coding model, 4-bit, 16K context | about 25.6 GB | 1 x RTX A6000 (48 GB) | $0.69 |
| A 70B model, 4-bit, 8K context | about 47.3 GB | 1 x A100 PCIe (80 GB) | $2.09 |
| The same 70B-class job on a faster card | about 47.3 GB | 1 x H100 SXM (80 GB) | $4.39 |
| Llama 3.1 405B, 4-bit, 8K context | about 256 GB | 1 x B300 (288 GB) | $8.99 |
| DeepSeek V3, 4-bit, 8K context | about 417 GB | 2 x B300 (576 GB) | $17.98 |
Time adds up. You pay the hourly rate for every hour the server is on, whether or not it is busy. So the habit is: start it when you are ready, work, download your files, and terminate it. Set a timer.
What an hour of muscle gets you
The useful question is not what the machine costs. It is how much work an hour buys. Pick a server setup and a model. The calculator checks whether the model fits, works out a rough top speed for one request at a time, and does the arithmetic. The speeds are estimates scaled from one measurement on our own home RTX 5090 (see the numbers above). They are not benchmarks, and the image time is your own number. Real speed depends on the model, the card, the software and how many requests the server answers at once, so replace the estimate with your own number when you have one.
Words from a model
Careful: every measured speed here came from our home RTX 5090, not from a rented server, and the estimate for any other card is scaled from that one measurement. The cost per million tokens below uses the price of the setup you picked and the speed in the box, so it only means something if your setup really runs at about that speed.
How the speed estimate works, and where its numbers come from
To write one token, a model of the kind listed here reads all of its weights from the card's memory once. So the memory speed the maker publishes for a card, divided by the size of the weights, gives a ceiling for one request at a time.
We have one measurement to compare with that ceiling: a 27B model at 6-bit on our home RTX 5090, one request at a time, no speed-up tricks. NVIDIA lists that card's memory speed as 1,792 GB a second. This calculator counts a 27B model at 6-bit as about 22 GB of weights (a calculated size, not the measured file), which puts the ceiling at about 80 tokens a second. We measured 42.6 with a short prompt and 39.3 with a 24.8k-token prompt: about 52% and about 48% of the ceiling.
The estimate applies the short-prompt share to the card and model you pick, and rounds down. That stretches one measurement a long way. It ignores the card's computing speed, the software and the length of the prompt, so real speed can be lower, most of all for a small model on a big card. With more than one card, speed depends on the connection between the cards, and we do not estimate it. The measured speeds with drafting (a speed-up trick) are offered only for the one card and model they came from.
Memory speeds used, as NVIDIA publishes them: RTX 5090, 1,792 GB a second; RTX A6000, 768; A100 PCIe 80 GB, 1,935; H100 SXM, 3,350; H200 SXM, 4,800; B300, up to 8,000. Whether a model fits uses this site's VRAM calculator at an 8K-token context; the models are counted at its 4-bit setting, except the 27B one at 6-bit.
Images, for example game characters
We have no measured image speeds for these cards, so nothing here is estimated and the setup above only changes the price. The 10 is an example to show the arithmetic, not a measurement. Replace it with a time you measured.
What to notice: the same machine at a higher speed makes each million tokens much cheaper. Keeping a rented server busy, with several helpers sharing one model, is the whole case for renting big. An idle server is the opposite: you pay the same rate for nothing. This calculator covers one request at a time only; we have not measured several at once.
A token is a piece of a word; one token is roughly three quarters of an English word. A page is counted as 500 words. A "game character" here means one generated image; a finished, usable character takes several tries, so treat the number as an upper bound. These are rough conversions, not measurements.
We will replace the estimated speeds and the example image time with measured ones, with the model, card, date and bill shown. Until then, treat them as a way to do the maths, not as a promise.
4. The honest limits
- Our real numbers come from one project on a higher-tier plan. We have not run the exact $20-plan version. The memory sizes in the table above come from our calculator, and the prices from RunPod's rates. How many helpers one machine can keep busy is not measured yet.
- Big open models are not the same as the best hosted ones. They can fall short on some tasks. Try yours on a small piece of real work before you hand it a project.
- The setup is real work. Your assistant can do much of the typing, but expect some trial and error.
- Check your plan's terms. Some plans limit how they can be used with automated tools or by many agents at once. Read yours before you build a workflow around it.
- The meter keeps running. Terminate the server when you finish, and download what you need first. A stopped pod still charges for its disk [2].
- Pod web addresses are public. If you expose a model or an agent dashboard, put a password on it [3].
5. A starter message for your assistant
Paste this into the assistant on your plan. It asks for a split of the work and a price, and it is told not to spend anything without your yes.
Plan the split: planner on my plan, muscle on a rented GPU
I have an AI assistant plan at about $20 a month and I want to build: [describe your project]. What I have: [iPad / laptop / PC with a GPU of about N GB]. Please help me split the work. 1. What you (the planner) should do with my plan's limited allowance: planning, review, hard decisions. 2. What a rented GPU running a big open model, or several helpers, could do in bulk. 3. Recommend the cheapest RunPod GPU setup that can hold a suitable open coding model, its hourly price, and an honest estimate of how many hours the work might take. Say which numbers are estimates and what you cannot know. 4. Tell me if renting is not worth it for this project. 5. Do not create, start or pay for anything until I reply "yes, go". When we finish, help me download my files, terminate everything, and confirm nothing is still billing. If you are not sure about a fact, say so instead of guessing.
You should get: a written plan with a price and an honest "is this worth it". Check: it names a GPU, an hourly price and the hours it assumes, and it waits for your "yes, go".
Ready to try the muscle side?
Make a RunPod account and add a small balance such as $10, or paste this into your AI assistant: Read agents.runpod.io and set up Runpod for me. Then use the starter message above.
Referral link: we may earn credits if you sign up.
Sources
- RunPod docs, Integrate your applications (OpenAI-compatible vLLM endpoints): docs.runpod.io/integrations/overview
- RunPod docs, Pod pricing and storage costs: docs.runpod.io/pods/pricing
- RunPod docs, Expose ports (the HTTP address is public): docs.runpod.io/pods/configuration/expose-ports
GPU rates are RunPod on-demand rates from its API, fetched within the last 30 days, and may have changed. Memory figures are estimates from this site's VRAM calculator. Hourly figures are arithmetic on those rates. The "$20 plan" is an example, not a quote; plan prices and limits vary.