Runpod•6mo ago

How to add more RAM to the Pod

Hi! Is there a way to set RAM manually? I mean, the current options is limited for example max 283gb RAM. But I need 1TB. How do I add more RAM?

Thank you.

Jason•7/21/25, 8:47 AM

use clusters with more GPUs

Jason•7/21/25, 8:47 AM

or cpu pods, with larger ram

JJason use clusters with more GPUs

akiratoya13OP•7/21/25, 11:05 AM

Is this what you mean? It's empty don't know why..

CPU pods only limited to about 250 gb RAM.

Jason•7/21/25, 11:46 AM

they have larger vram because they have more gpu's yes

Jason•7/21/25, 11:46 AM

(in one cluster)

Jason•7/21/25, 11:47 AM

btw clusters are expensive tho

Jason•7/21/25, 11:48 AM

nvm you can go here (look below in my screenshot) and increase more gpus and still get 1tb of ram

Jason•7/21/25, 11:49 AM

These setups allows you to get 1tb of ram

Jason•7/21/25, 11:49 AM

these are in pods

JJason These setups allows you to get 1tb of ram

akiratoya13OP•7/21/25, 2:47 PM

Ah okayy, finally understand the parameter for it. Thank you so much..

So.... with about $7/hour, I can try use kimi k2 for my self? hahahaha. And try the quantized version with so much cheaper price? :D:D

Anyway thank youu

Jason•7/21/25, 2:50 PM

Hmm or use the amd ones

Jason•7/21/25, 2:51 PM

Try apis if you don't wanna self host just for your self (if it's cheaper that way why not)

JJason Try apis if you don't wanna self host just for your self (if it's cheaper that w...

akiratoya13OP•7/22/25, 3:43 AM

apis? Do you mean like OpenRouter, Chutes?

Jason•7/22/25, 4:23 AM

Ye openrtr tgther ai

Jason•7/22/25, 4:24 AM

Unless you need high throughput, that is effecient on cloud gpus

JJason Unless you need high throughput, that is effecient on cloud gpus

akiratoya13OP•7/22/25, 11:10 AM

I see... I thought of using Runpod because of hit rate limit on free one. and if paying, I wonder if assumed 1 hour on OpenRouter kimi usage can be cheaper than I use runpod vs speed. Basically it gets back to price value.. If runpod is cheaper for 1 hour full usage then it's really worth it ( based on my tests a couple of days and for my use cases ).

What do you think?

akiratoya13OP•7/22/25, 11:11 AM

If you said openrouter will be cheaper then I won't bother trying since it's also kind of a work to make the template first for gguf one. << Never try this, I also want to know if gguf one (from unsloth) is enough for roocode usage.

Jason•7/22/25, 11:27 AM

Hmm depends on what you use

Jason•7/22/25, 11:27 AM

How much tokens estimate

Jason•7/22/25, 11:28 AM

Most likely if you're alone, using the full kimi k2 it's more Exp in runpod or wherever it is, paying more than. .8 per hr

JJason How much tokens estimate

akiratoya13OP•7/22/25, 11:42 AM

wow... This is really a hard question hahaahah. Honestly I don't know, but I assume it is actually so much because when I try using gemini pro, it's taking maybe about 80M token within 3 hours, so it's around 28M token per hour or more depending on the task.

I see that are cases that is not using much token, but there is using so much token that I'm not even except it is possible.

akiratoya13OP•7/22/25, 11:43 AM

Based on your comment, I think it's wise for me to just try using runpod for full 1 hour paying to see how much it is ( I don't want to do this at first ). hahahah. But still, on OpenRouter, total token matter so much when runpod is no, that's also some things to consider. If we're using maximum token fully 1 hour, of course runpod will be cheaper right?
I need maybe 50t/s since when I use roocode, I got that speed ( based on what chutes provider said on openrouter )

Aakiratoya13 Based on your comment, I think it's wise for me to just try using runpod for ful...

Jason•7/22/25, 11:54 AM

The runpod costs are estimatable if you're using pods, even without using it

Jason•7/22/25, 11:54 AM

Just multiply (the price per hour + cost of your storage/hr) with your usage

Aakiratoya13 wow... This is really a hard question hahaahah. Honestly I don't know, but I ass...

Jason•7/22/25, 11:55 AM

Wow that's quite heavy use

JJason Wow that's quite heavy use

akiratoya13OP•7/22/25, 11:56 AM

I also shocked by it.. I don't know how gemini cli works at that time, but it really burns the token so much.. hahaha.

Aakiratoya13 Based on your comment, I think it's wise for me to just try using runpod for ful...

Jason•7/22/25, 11:56 AM

Yeah sure try the perf on runpod

Jason•7/22/25, 11:57 AM

Oh gemini cli? Maybe it's agentic use

Jason•7/22/25, 11:57 AM

It explores many files, thinking

Jason•7/22/25, 11:57 AM

Tool calling

Jason•7/22/25, 11:57 AM

Not sure if kimi k2 can support cli like use

Jason•7/22/25, 11:57 AM

If it's not fine tuned for it

JJason Not sure if kimi k2 can support cli like use

akiratoya13OP•7/22/25, 11:59 AM

No no.. after I got billed about $35 within 4 days, I try roocode instantly, and try for the free, and honestly it's good. But about these 2 days, somehow I get limited so much ( maybe the chutes doing the limiting, since openrouter says 1000 request / day, so it should be okay but no ).

That's why, if I must pay, I think I want it to be no restriction and of course low cost since it's all for my personal research.

Jason•7/22/25, 12:01 PM

ahh ic

Jason•7/22/25, 12:01 PM

How much are you willing to pay?

Jason•7/22/25, 12:02 PM

my friend said he managed to run it in 4x or 5x mi300x (its low stock currently) with 4k context 50t/s, usually you'd want a higher context tho
if you're okay with ram offloading ig it fits with other gpus

JJason How much are you willing to pay?

akiratoya13OP•7/22/25, 12:05 PM

This is also hard to say since this is relative type question. Like if it's 50t/s, I feel this already pretty pretty good, I'm okay with paying $1.5/hour ( Note: I'm not like a crazy person that turns this on 24 hours HAHAHA. I use it when I'm only on computer too.. So maximum is 3 hours a day ). If 50t/s with 64K token, I already feel that's enough. Even if it's 30t/s, I also already okay.

So let say I can even get it on much much lower rate, I also okay with 15t/s

. Sorry to be like a cheapskate, but I'm not rich enough to burn money.

Is it too much to ask?

Jason•7/22/25, 12:06 PM

woah

Jason•7/22/25, 12:06 PM

1.5$ an hour?

Aakiratoya13 This is also hard to say since this is relative type question. Like if it's 50t/...

Henky!!•7/22/25, 12:06 PM

When I said 50t/s I mean processing

JJason my friend said he managed to run it in 4x or 5x mi300x (its low stock currently)...

akiratoya13OP•7/22/25, 12:07 PM

Depends on t/s. If it's acceptable, of course I'm okay offloading to ram. On my local PC, I try it with offloading, but I only can run up to 14B model Q4, and when using high context, I already got soooo poor speed.

Jason•7/22/25, 12:07 PM

so to run kimi you need around 1tb+ of mixed ram and vram?

Jason•7/22/25, 12:07 PM

is t hat correct

Henky!!•7/22/25, 12:08 PM

If you load a full 64K prompt at 50t/s pp it would take 21 mins

Henky!!•7/22/25, 12:08 PM

But currently testing how fast it is on nvidia, the 50 was on AMD

JJason so to run kimi you need around 1tb+ of mixed ram and vram?

akiratoya13OP•7/22/25, 12:08 PM

yeah hahahaah! That's why I want to try the quantized one by unsloth team ( the 280gb size one ).

Jason•7/22/25, 12:09 PM

i guess $1.5 nots going to run it (the full one)

Aakiratoya13 yeah hahahaah! That's why I want to try the quantized one by unsloth team ( the ...

Henky!!•7/22/25, 12:09 PM

Which quant is that?

Jason•7/22/25, 12:10 PM

How to add more RAM to the Pod

Similar Threads

Similar Threads

Similar Threads