Meet Number 6
Battlestar Galatica 2003 Remake
Ok, not that number six, this number 6.
This is my new AI server, named after Number 6 from one of the best sci-fi shows ever made Battlestar Galatica (2003 reboot). I tried to come up with a cool AI name that was based on a movie or show, then I thought about one of the best shows ever made.
What is this thing?
Technically, it's my old computer before my recent pc upgrade. I've been looking for a decent AI server that is local but fast enough that I don't want to throw it out of the window.
I had a lot of fun tinkering with the Strix Halo but it is just a toy and extremely overrated for any actual workload. It is great to tinker with and play with much larger models than you usually can.
Specs
AMD 5950X
Asus Dark Hero AM4 Motherboard
32G DDR4 Ram
Dual Nvidia RTX 6000 Pro 600W Workstation Edition
It doesn't look like much, until you see the GPUs. These are faster 5090's with a full 96GB vram each giving me a total of 192G vram to work with. This system was just sitting in the closet and would be a perfect choice for two RTX 6000 Pro with very minor performance loss off a top of the line AM5 system.
I did run into a few problems, the major one was the PSU did not fit in this case. It is considering larger than a typical PSU so I am currently sitting it on the desk behind the case. I needed a larger power supply as well as one that supports two 12V 2x6 cables.
I am planning on another H9 Flow case like my main system that will fit the larger PSU and give it even more airflow.
While 32G DDR4 ram is not impressive, I do not plan to do any CPU offloading and will only use the GPUs. I used to have more ram in this system but it isn't needed.
What am I running on it?
I am currently testing models and tweaking performance. I am currently running GLM 4.5 Air FP8 on SGLang while I wait for GLM 4.6 Air to be released.
GLM 4.5 Air is a 106B parameter model from Z.ai and considered the best model at this size. The model weights alone come in around 118GB and everything else like kv cache and context you can push 192G easily.
I have a few other models I want to test. I have been using GPT-OSS-120B locally for a while, I was able to get 50 tokens/sec on the Strix Halo, which is really good but it does really poorly under real usage due to the slow ram making prompt processing really slow.
While GLM 4.5 Air is a slightly smaller model 106B parameters compared to 120B of GPT-OSS-120B, it has 4x the active parameters when in use. These models are called MoE or Mixture of Experts models. They are really popular as you can get some amazing speeds due to only some of the parameters being active at a time. For GPT-OSS-120B, only 3B parameters are active when being used where GLM 4.5 Air uses 12B. I am also running GLM 4.5 Air at FP8 quantization which is twice as large as GPT-OSS-120B's MXFP4. In other words, GLM 4.5 Air is considerably more demanding to run.
I have been seeing as much as 138 tokens/second peak from this rig on GLM 4.5 Air, with most requests giving me 100-120 tokens/second. Even at 122K context, I am still seeing around 75 tokens/second. The prompt processing however is very high making it really quick to spit out the first token (ttft).
Here you can see it summarizing a book that represents around 127,00 tokens, very close to the maximum 131,072 tokens this model is capable of. As you fill up the context window with data, models get a lot slower. I am still able to reach an impressive 77 tokens per second at max context.
This thing is a beast, ideally I want to be running the full 357B parameter GLM 4.6 but until DDR6 is released, I will stick with this two GPU setup.
Power usage
Here is where things get interesting. To summarize that book with 127K tokens trips my UPS at almost 1400W draw.
The PSU can handle 1500W, but there are other things on it.
If I power limit the two GPUS to 300W, I can drastically reduce the power draw down to 784W
Surely this means I am getting around half the speed?
I lost a whole 3 tokens/second! I lost about 3.9% in performance but reduced power usage by 43.43%!
Quite a fair trade I must say. Nvidia does make a model of the card called MaxQ that are fixed to 300W total and have a different fan style. I didn't want to pay the same amount for a less potential power if I decide to use them differently.
Why?
Good question. My primary reason is for analyzing stock data. I do not believe AI can predict price action, they can however churn through a massive amount of data and if driven properly can increase your edge or alpha. I already heavily use AI for trading, much of which is using cloud providers, but I don't want my data going to third parties.
You would be surprised how interesting many of us are after finding your last two posts about your set up
I like to post updates when I change things.
🔥💀 HIVE JUST HIT $0.11 — SINGLE DIGITS LOADING!!! 🚨📉💣 💀🔥
https://hive.blog/hive-167922/@networkallstar/hive-just-hit-usd0-11-single-digits-loading
Nice
So much of this was over my head, but it sounds pretty cool and I am excited for you.
😄 mostly the AI stuff. The hardware stuff I get.
Hive is done
https://hive.blog/hive-108278/@kgakakillerg/hive-is-done
That power saving is massive for the performance loss damn! Hope energy is pretty cheap there, or you got solar to help, I'd cry looking at my bill with that kinda power usage haha, it's still fairly bad here in the UK.
This post has also reminded me that I need to fetch and watch battlestar galactica!
power is crazy here, 0.256c KWH, it has gone up from 0.15c years ago.
I keep thinking about solar, but I hate that it takes 20 years to break even and by then you need to replace it. Plus I live with snow, so I got to pay someone every year to clean the snow off in bad storms. I am thinking about putting some off to the side though.
The reboot is sooooo good. Some of the best TV you will see.
I think the same a lot of the time, that and being able to afford it in the first place. But if I could afford it I'd probably do it on a moral basis and kinda ignore the ROI prospects. If it could manage 70% to breaking even I'D be fine with that honestly. Though in a few years time when the newest tech gets more affordable this could change.
Caprica was a better show than BSG. I really enjoyed where it was going with the themes of Post-Humanism in the lead up to number 6. I am still sad it was cancelled. One of my favourite shows, but the BSG reboot is close.
When do you get a rambling hybrid installed in a bathtub in the spareroom for more organic tokens per second?
I forget if I got around to watching Caprica. I plan on rewatching the series soon, I almost never re-watch shows. I'll add it to the list.
So glad these GPUs arrived for you, and weren't part of the cargo plane crash :)
We can get up to 2400W at the wall here, but I do not have the funds to obtain 3 of these cards :P
I was honestly worried they wouldn't show up.
🔥💀 HIVE JUST HIT $0.11 — SINGLE DIGITS LOADING!!! 🚨📉💣 💀🔥
https://hive.blog/hive-167922/@networkallstar/hive-just-hit-usd0-11-single-digits-loading
😂😂😂😂😂😂🗑️🚮
@themarkymark @buildawhale @acidyo @ocdb @azircon @crimsonclad @gogreenbuddy
You are a scammer farmer racist downvoter 😂😂😂😂😂
You really think anyone cares about rep on Hive 😂😂😂😂😂
Just shows how childish you are 😂😂😂😂😂😂
If you didn't know I've been getting downvoted for a long time on Hive
They have been trying to bully me with downvotes and words but it doesn't affect me I've been online for over 25 years
So I've dealt with people like this before it's nothing new
But what they are doing is bad Very badi joined steemit like everyone else thinking it offers freedom but I must say steemit offered much more freedom than Hive does
Everyone already knows I only post my own original content I have no need to steal anything from anyone
I've been labelled a lot of things on Hive 😂 by the biggest farmers scammers and downvoters
Actions speak louder than any words do especially when them words are coming from a compolsive liars
I will keep updating this post when I get time
https://hive.blog/hive/@ureka.stats/the-untrending-report-hive-downvote-analysis-16-09-2025-20250916181314
https://hive.blog/hivedev/@makerhacks/please-peakd-allow-me-to-completely-block-spammers-would-you-ever-vote-for-this
https://hive.blog/hive-167547/@dlmmqb/re-kgakakillerg-t34wy6
https://hive.blog/hive-170475/@acidyo/re-kgakakillerg-t32sw1
https://hive.blog/hive-167922/@themarkymark/re-kgakakillerg-t30ziu
https://hive.blog/hive/@ureka.stats/the-untrending-report-hive-downvote-analysis-29-09-2025-20250929172045
No downvotes for this racist person 🧐
https://hive.blog/hive-196387/@acidyo/re-kgakakillerg-t41yr6
https://hive.blog/dvs/@dhedge/re-kgakakillerg-t3clz3
https://hive.blog/hive-150329/@kgakakillerg/t3a8fu
https://hive.blog/musical/@steevc/re-kgakakillerg-t2kz1m
https://hive.blog/hive-108278/@kgakakillerg/hive-is-done-part-2
https://hive.blog/hive-121744/@ocdb/re-brosonleo-t11mdc
Particularly regarding financial and business decisions. It has been a weakness of the financial system that a variety of gatekeepers (lenders, brokers, factors, and etc.) necessarily attained to strategic and other proprietary information in order to facilitate operations. That has been enormously improved by digitization in many ways, but using cloud services dramatically worsens that data insecurity.
It has been one of the most alarming features to me of the surveillance state that has arisen that business information, including metadata regarding principals, their plans, key hurdles, and etc. are all harvested and available to an assortment of analysts and parties with such execrable ethical and moral standards they'd work in that industry. The notorious collection of >16M penis pics by one such analyst well characterizes that ilk, and sensitive business information being at their fingertips is ill-advised, IMHO. I am actually stunned that we aren't inundated with reports of people being ruined, lucrative trades based on insider information, and massive wealth concentration in the wallets of analysts on the daily. OTOH, the lack of such reports doesn't indicate the lack of such swindles, given the nature of that beast.
The more you know, the less you want others to know what you know.
That's a pretty impressive AI setup. I'm sure you have equally impressive uses for it.
Thanks!
Watched both BSG shows with my son. Does AI really help your trading?
Yes
😂😂😂😂😂😂🗑️🚮
@themarkymark @buildawhale @acidyo @ocdb @azircon @crimsonclad @gogreenbuddy
You are a scammer farmer racist downvoter 😂😂😂😂😂
You really think anyone cares about rep on Hive 😂😂😂😂😂
Just shows how childish you are 😂😂😂😂😂😂
If you didn't know I've been getting downvoted for a long time on Hive
They have been trying to bully me with downvotes and words but it doesn't affect me I've been online for over 25 years
So I've dealt with people like this before it's nothing new
But what they are doing is bad Very badi joined steemit like everyone else thinking it offers freedom but I must say steemit offered much more freedom than Hive does
Everyone already knows I only post my own original content I have no need to steal anything from anyone
I've been labelled a lot of things on Hive 😂 by the biggest farmers scammers and downvoters
Actions speak louder than any words do especially when them words are coming from a compolsive liars
I will keep updating this post when I get time
https://hive.blog/hive/@ureka.stats/the-untrending-report-hive-downvote-analysis-16-09-2025-20250916181314
https://hive.blog/hivedev/@makerhacks/please-peakd-allow-me-to-completely-block-spammers-would-you-ever-vote-for-this
https://hive.blog/hive-167547/@dlmmqb/re-kgakakillerg-t34wy6
https://hive.blog/hive-170475/@acidyo/re-kgakakillerg-t32sw1
https://hive.blog/hive-167922/@themarkymark/re-kgakakillerg-t30ziu
https://hive.blog/hive/@ureka.stats/the-untrending-report-hive-downvote-analysis-29-09-2025-20250929172045
No downvotes for this racist person 🧐
https://hive.blog/hive-196387/@acidyo/re-kgakakillerg-t41yr6
https://hive.blog/dvs/@dhedge/re-kgakakillerg-t3clz3
https://hive.blog/hive-150329/@kgakakillerg/t3a8fu
https://hive.blog/musical/@steevc/re-kgakakillerg-t2kz1m
https://hive.blog/hive-108278/@kgakakillerg/hive-is-done-part-2
https://hive.blog/hive-121744/@ocdb/re-brosonleo-t11mdc
No one cares 😂😂😂😂😂😂😂😂😂😂😂😂😂
😂😂😂😂😂😂🗑️🚮
@themarkymark @buildawhale @acidyo @ocdb @azircon @crimsonclad @gogreenbuddy
You are a scammer farmer racist downvoter 😂😂😂😂😂
You really think anyone cares about rep on Hive 😂😂😂😂😂
Just shows how childish you are 😂😂😂😂😂😂
If you didn't know I've been getting downvoted for a long time on Hive
They have been trying to bully me with downvotes and words but it doesn't affect me I've been online for over 25 years
So I've dealt with people like this before it's nothing new
But what they are doing is bad Very badi joined steemit like everyone else thinking it offers freedom but I must say steemit offered much more freedom than Hive does
Everyone already knows I only post my own original content I have no need to steal anything from anyone
I've been labelled a lot of things on Hive 😂 by the biggest farmers scammers and downvoters
Actions speak louder than any words do especially when them words are coming from a compolsive liars
I will keep updating this post when I get time
https://hive.blog/hive/@ureka.stats/the-untrending-report-hive-downvote-analysis-16-09-2025-20250916181314
https://hive.blog/hivedev/@makerhacks/please-peakd-allow-me-to-completely-block-spammers-would-you-ever-vote-for-this
https://hive.blog/hive-167547/@dlmmqb/re-kgakakillerg-t34wy6
https://hive.blog/hive-170475/@acidyo/re-kgakakillerg-t32sw1
https://hive.blog/hive-167922/@themarkymark/re-kgakakillerg-t30ziu
https://hive.blog/hive/@ureka.stats/the-untrending-report-hive-downvote-analysis-29-09-2025-20250929172045
No downvotes for this racist person 🧐
https://hive.blog/hive-196387/@acidyo/re-kgakakillerg-t41yr6
https://hive.blog/dvs/@dhedge/re-kgakakillerg-t3clz3
https://hive.blog/hive-150329/@kgakakillerg/t3a8fu
https://hive.blog/musical/@steevc/re-kgakakillerg-t2kz1m
https://hive.blog/hive-108278/@kgakakillerg/hive-is-done-part-2
https://hive.blog/hive-121744/@ocdb/re-brosonleo-t11mdc
Wow
And you really trust Ai for your trading??
seems to be working
Okay. Great.
And stop downvoting me please
I don’t know what to do
Hm i guess you use ai it for crypto / market trading, but it could be useful for sport trading on exchanges 🤔
Amigo me puedes decir porque le diste voto negativo a mi publicación sin explicación alguna
As many of you know I've been getting downvoted for a long time on Hive 😂 😂 😂 😂 😂
Even though I create my own original content daily
The one's doing the downvoting are the biggest farmers and scammers on Hive 😂 😂
Hive is done
https://hive.blog/hive-108278/@kgakakillerg/hive-is-done
Hello @themarkymark! I've been following your experiments—they're impressive. It's like a real saga about creating your own AI.
It would be interesting to see the final result of your work, or at least an intermediate one :). I'm an experimenter myself, but in the field of chemistry, especially molecular design in organic and biochemistry... Though for now, I write in a simplified way—people don't really like reading scientific texts 😀. But in popularizing science, this makes a lot of sense... and there are results.
However, I've run into a problem. It seems I'm on the blacklists of @buildawhale and @freebornsociety, and they downvote all my posts. A long time ago, I was on the blacklist of hivewatchers. Now, I've been off their list for a long time.
Is there any way to resolve this issue? Could you help me? I really don't want to abandon running this blog. I want to attract more people to the Hive project from my field, but I don't want to deceive them. After all, I'm under "sanctions" from the whales :))). And I definitely can't compete with them, nor do I see the point. I believe they should exist so that the whole system doesn't turn into an info-dump. I think Hive has potential. Please take a look at my publications. I have no intention of using the Hive platform dishonestly. I don't need that. Yes, I've had breaks in blogging—I was on an internship, and there wasn't enough time (lab work is a special topic—it takes up a lot of time :). But that's probably not the reason...
I really hope for your help or at least a recommendation on what can be done in this case.
I would be grateful if you find the time to reply to me.
I'm eagerly awaiting your response.
Thank you for such a quick response.
Absolutely correct. I do not deny the fact of my mistake and I admit it. But a lot of time has already passed. I realize my mistake. I was a newbie on this platform and was just figuring out how everything works. And I understand that AI generation is bad. After that, I contacted "hivewatchers" and we agreed via Discord that if I make 30 quality posts, I can prove that I have no intention of using Hive for manipulations. Which I did. After which I was removed from the blacklist.
However, I want to note that after I was put on the blacklist, I decided to check myself again. That is, my popular science texts that I wrote 100% myself. And surprisingly, the check showed that I used artificial intelligence. But that was not the case at all. I had to write only in the first person, and only then the AI percentage was significantly reduced. Although writing in the third person about something is much more interesting and natural. That is, these systems are not perfect. A lot is written about this...
That's the story. My last hope is for your help.
Hive is done
https://hive.blog/hive-108278/@kgakakillerg/hive-is-done
Original content, that you created, every time, or it gets the flag.
I don't know why baw is flagging you, but I'd suspect you are passing off content that wasn't created by you.
IF that is the case, make original content, every time, or you will never get rewards.
OR, you can set rewards to burn and post as you please.
Doing this may get you more readers that you can monetize in the comments or later posts.
Thank you for your response. But I've already created many publications, and unfortunately, all of them are being blocked. Among them, I made 30 publications where I renounced the reward to prove that I'm not manipulating. Unfortunately, that didn't help. If I even relied on some information, I always indicated the primary source. If the last publications that remain before the payout get downvoted, then I see no point in continuing this project. Unfortunately, I'll have to shut it down. Negative experience is also useful. I'll look for other platforms. Maybe I'll find something, or create my own.
The only thing I'll do from this account is that it's interesting for me to follow the experiments of @themarkymark and maybe support someone who was lucky enough not to fall out of favor with the whales, at least a little.
I have removed you from ZERO, but further AI will result in remaining on the list.
Thank you so much for your understanding. I promise it’ll be only original content. 🫡 We’ll do our best to make Hive interesting...
Hive is done
https://hive.blog/hive-108278/@kgakakillerg/hive-is-done
Hive is done
https://hive.blog/hive-108278/@kgakakillerg/hive-is-done
Set your rewards to burn and see if that helps.
Ok. I'll try to do it that way, maybe it will help. I'll experiment.
Nice one 👌