Skip to content
PodSubs
ShowsPeopleGuestsJobsNewsEvents
More
BlogPodCred
Menu
ShowsPeopleGuestsJobsNewsEventsBlogPodCredNetworksSign in
Join nowSign in
  • Shows
  • People
  • Jobs
  • News
  • More
    • Guests
    • Events
    • Blog
    • PodCred
    • Networks
    Join nowSign in

Mere Morpheus

Mere Morpheus artwork

Setup The Morpheus Skill | Agentic Compute Tutorial

Apr 28, 202624 minAudio
ShareXLinkedInFacebookWhatsAppEmail

I got inference coming out at all ends! Welcome everyone, Kyrin here with Ep#20 of Mere Morpheus. Today we're going through a practical, start‑to‑finish setup of the Morpheus Skill to unlock Claude‑quality inference at a fraction of the cost. I cover the key prerequisites of having an agent framework running (I’m using OpenClaw), being able to talk to an LLM already and holding some MOR tokens to stake. From there I demo staking on the mor.org Builders Dashboard, jump into app.mor.org to create an API key before getting the last command from morpheusskill.com to get your new access to inference up and running. This is a Value 4 Value show so please provide back any value you got with some time, talent or treasure! Extra links: mor.org app.mor.org morpheusskill.com Timeline: (00:00:00) Intro (00:00:33) What is the Morpheus skill and why use it? (00:02:23) Prerequisites: agents, LLM access and tokens (00:04:37) Staking on mor.org and builder dashboards (00:08:26) App.mor.org: API keys, models, pricing and usage stats (00:13:27) Connecting the API to your agent via Morpheus Skill (00:17:21) Live demo setup: OpenClaw, models and gateways (00:20:41) Verifying billing changes during the demo (00:22:49) Subnets on Base and Arbitrum; wrap‑up If you would like to support me for the value I provide you can use the referral link below when depositing capital and we will both earn bonus MOR. Alternatively you could send me a token of appreciation directly to my wallet ;) Referral Link: https://dashboard.mor.org/capital?referrer=0xE935f231c99c04Ee0b4532a3d0BdA81B152a0384 Wallet Address: 0xE935f231c99c04Ee0b4532a3d0BdA81B152a0384 Connect with Mere Mortals: Website: https://www.meremortalspodcasts.com/ Discord: https://discord.gg/dX9BEhT3kz Twitter/X: https://twitter.com/meremortalspods Instagram: https://www.instagram.com/meremortalspodcasts/ TikTok: https://www.tiktok.com/@meremortalspodcasts Value 4 Value Support: Website: https://www.meremortalspodcasts.com/support Paypal: https://www.paypal.com/paypalme/meremortalspodcast

Audio and video play straight from the show's own hosting. PodSubs stores and serves no media.

Transcript

From the show. Tap a time to play from there.

Kyrin DownI got inference coming out at all ends. Welcome everyone to another episode of the mere Morpheus podcast. I'm your host here, Kyren, live on the 04/28/2026. And today, we're gonna be doing a bit of a tutorial. Indeed, we're gonna be trying to set up the Morpheus skill. Now I actually have set it up, but I'm gonna be doing a slight run through. So showing you a couple of the different steps. We'll go over some prerequisites of what I would recommend already having before getting started,

before jumping onto a couple of different websites you need to connect to, Before finally, we'll try a bit of a live demo showcasing jumping from different models and how the billing process works and maybe even how you can get yourself $10 of free credits. So let's just jump on to, I guess, what some prerequisites are and what the actual Morpheus skill is. So so the the Morpheus skill, for those who don't know, is a

Claude quality inference at 78% less cost. So it's essentially using your Morpheus tokens to be able to get compute credits, and then using that via I in in my case, I'm using it via an agent, in in particular OpenClaw. But I imagine you can do this via another sort of interface. I haven't dived into all of the myriad of ways that you can use this inference. But certainly, if you're more technical, you could go down and integrate it directly into, like, the back end of an app or something like this.

Or yeah. If you're more basic like me, you're just really trying to start things out and see, like, how can I actually get to the point where I can use this inference to talk to an LLM? And then from there, we can do all of the zany apps and crazy things beyond that. So we'll we'll perhaps talk a little bit more in future episodes about pricing, about tiers, about different models, and why it is or is not, depending on your circumstances, a sound economic decision to actually have the Morpheus inference, and some of the trade offs and benefits that actually come from having this. So what are a couple of prerequisites? Well, I would say you probably already want to have an agent framework or some sort of setup

already in place. So in particular, this is going through what I was talking about recently in past episodes of the one command click easy install not being true. Get through the hard work of actually getting that done, whether you're on a Mac Mini, whether you are doing it through Windows Subsystem for Linux like I am, or whether you're a full blown Linux user. I would also say that you probably wanna be at the point where you're already talking to an LLM

kind of via your agent or through your agent so that you've already got some sort of inference or compute already available. You know, whether this be I did it through getting Venice first because the Morpheus skill on top of that is yeah. I'm not sure. It it was actually exceedingly easy. Maybe if I'd done it the other way around, doing the Morpheus skill would have been hard and then getting Venice would have been exceedingly easy. I think it's just once you get to a stage where you already

have something that you can talk to and can kind of it debug itself, that's the point you wanna be getting to when you're starting to add more models in because it just makes it so much easier. So, you know, perhaps, like, their most easiest route is having some OpenAI or a subscription to some place, get that set up, and use it to get you set up via the more decentralized options. And then it's a lot more easier. At least that's probably what I would suggest.

And you probably already want to have bought some Morpheus of Venice. So you already want to know how to, like, go onto base or Arbitrum and be able to buy tokens and, you know, be familiar with connecting a wallet to a to a a dashboard or things like this. That's probably all also the a prerequisite, I I would say, for for where you wanna be at. So let's jump into, I guess, some of the steps and how you would actually go through doing the Morpheus skill and getting that connected to your agent. So gonna jump over to here.

And for those who are just listening via the audio, there is a video of this as well which would probably make it a lot more easier. I would suggest starting off with more.org and staking to the subnet that is required to be part of the Morpheus inference marketplace. So if you're completely brand new, don't even know what Morpheus is, you know, come here, you can click the buy more button, and that'll show you kind of like what the contract is, what the price is, and then you would have to go on to like a decks of some sort and do that yourself.

Once you actually already have some Morpheus in a wallet, something like a Rabi MetaMask, something like that, you can see I've got some Morpheus in here. We then go over to the dashboards and the builders dashboard. This will take you to a section here listing out all of the dash builders that are available. So I actually have the Miamorpheus way down in the list somewhere here where I've got some of that staking to myself. There we are, Miamorpheus.

At the top here is where you can see, because this is pretty popular in terms of more staked, it's almost at a 100,000. The Morpheus Marketplace API, I will just show that I'm already staking in it. And the reason I would recommend doing this first is because it takes kinda twenty four hours potentially to kick over to actually get your stake. So depends on when you're staking, but the a lot of these places will do like and Venice does the same as UTC midnight,

which is 10AM Australian Standard Time or Australian Eastern Standard Time, is when it kind of like this daily inference ticks over. So it does take, like, potentially, you know, twelve hours if you're doing it at midday UTC for it to kick over. So that's why I would suggest doing this first. So this is where you connect your wallet to the website. You can then just come over to here, go to the Morpheus Marketplace API. This is on base. There is another one for Arbitrum.

And this is where you just stake a certain amount of Morpheus, however much you want to it. I did 40 just as a test here, and I will probably be doing a little bit more later. And you can go down the list here and see all the various different stakers. If I went through quite a whole bunch of them and got to page 10 or something, then you would you would see myself. At the moment, there's only 60 of us or 60 different wallets. There could be multiple different users staking for different purposes. Lock period of seven days. And the description here, the official subnet for Morpheus

Marketplace API, the subnet will serve as the message to stake for access. Staking within the subnet will accrue daily credits within app.more.org. So that's what you want to be doing, staking to it the at the moment. And basically, how it works is of all of the total Morpheus staked here, you will get a percentage of credits based on how much is in there. So me being the tiny shrimp I am at 40 more, I

believe at the moment I get 0.08¢ of credits daily, and that will vary depending on how much is actually staked. If more people stake, that'll be less. If I stake more, it'll be more. If some people drop out, it'll be more. And it's very confusing saying more and Morpheus because the token is called more. Very good current. So that's the the first step that you wanna do. Go to more.org, go to the dashboard, get your builder set up, and then you are in here. Now it also mentions here app.more.org.

So that is the next place you go to. So if you click on that, that'll take you to here where it should jump, I guess, straight to your API keys. No. It actually won't because I'm already logged in. So you'll need to log in first just via a simple email, and then you get into this area here. So we see we've got API keys, billing, models, pricing, usage, analytics, and you can do a little test, and then finally, documentation. So you can see I've already got my API key set up and I created it yesterday.

And this will create a key that's called s k, a bunch of different digits and numbers. If we go into the other sections here showing the billing, you can see this area here is where it finally showed me my credit. So because I did this yesterday at, like, 10:30AM, I I had to wait the full twenty four hours. So a little a little annoying because you wanna get stuck in and started straight away, but that's why I recommend doing the Morpheus

staking first and then jumping over to this area. So you connect your wallet that you use to stake your more. You can see down here I've got my forty forty more staked here. You can also, if you want to, just buy credits directly within the app via crypto or credit cards. I've got 0.08 daily, and I've already used 0.06 or 0.07, hence why I only have 1¢ of daily credit remaining. Because they're kinda trying to bootstrap this, you do get a little bit of bonus cash here. So you got $10 of credits.

And when we do our live run through, I'm going to try and test out and see what happens. But basically, it should use your daily a d API credits remaining before jumping over onto any available credits. So let's say that you've got, you know, $5 worth or a dollar's worth or like myself, 8¢ worth. It should use this up daily before then jumping over to your kind of like backup system that you've got over here. So that's a a nice little way of kind of

restricting yourself or just monitoring being like, alright, I've got these kind of daily credits and I've got a little back up stash. You can almost think of it as like a I don't know, using a a debit card on on the regular. And if you get that down to zero, you've got an automatic system that'll kick in and go to another card where you perhaps have some savings or something like that. So that's the billing. Shows all the different models here that are available and what they are, what they do,

and you can jump further into that. And the model pricing as well, which I found really useful because it gives you just, like, a little bit of a rough guide as to which models actually cost the most and why. So and what like, what they're doing. So you've got GLM 5.1, Qumikay 2.5. So I was using this one as my default on Venice for ages, and it was only until I kinda came here and I saw like, oh, there's actually a couple of models that are much cheaper, the Gemma four thirty one b

or the one that I switched to recently, which was, if I find it here, GPT OSS one twenty b, and it's only got 0.07¢. Now, obviously, all the different models give you a different output of variety of, you know, if you're the complex task that you're doing, what it's meant to be doing. So you can have advanced chat and content, got something that's classification simple, Mistral basic agent functionality,

Venice Uncensored for the creative use, and ones which are more suited for code development, ones which are cost efficient AI agents with the mini max. All these sorts of things, you probably need to test around a lot. I have not done a lot of testing, so that'll probably also be a future episode as well. Finally, usage analytics. You can kinda see how you're using your credits and what they've been using on. So I was just playing around a little bit this morning, and you can see I used my

Morpheus inference here. What did I spend it on? It was spent on GLM 4.7 flash, even though I have a couple of other options which I was playing around with as well. And, yeah, you can dive into it further and further, like credit spend, how much your stake spend, daily token volume, understanding usage, all these sorts of things. For those who are data geeks and love it, you can filter by date, time, all that sort of stuff and export the data as well.

So you've got your Morpheus tokens, you've staked them, you've now gone to your dashboard and you've gotten yourself an API key. What is the next step? How do I connect this to my agent? And this is where we go to the Morpheus skill website, which is something David Johnston has set up. So jumping over to here, once again, morpheusskill.com. And this time, it actually was a relatively easy one click install.

So kudos to David. Once you've actually got some stuff set up, it it actually was a lot easier. So, you know, what do you get? 30 plus open source models. You're owning your own inference. You can use this on any agent platform. So OpenClore, Homey's agent, IonClore, Nanoclore, any platform that supports agentskills.io. And, you know, the big whole reason you wanna use Morpheus in the first place is because it's like the peer to peer marketplace, falls back to local

llama, no API bills, no credit limits, things like this. So it's really you're not gonna get cut off at the source. The only source is how much capital you're willing to put into it to to get out there. So I can give you a little show through of what I actually did yesterday. So I simply just copied this curl, the FSSL h get everclaw.xyz because it used to be called Everclaw, and he's just renamed it to the Morpheus skill for a little bit more easy ease of use. And he shows here the different things that you can use, and it's all the same curl. So

I tried this out myself, and here's just what actually happened. So I copied that, pasted that into my terminal here. So I would've literally, if if you wanna see me do it real time, just copy this, copy it, go over to here, put it into your, in my case, the Ubuntu setup, and then I just run that command. This is what I actually got whilst running the command. And there are still some, like, minor bugs here and there, for example. So

it did a lot of the stuff at the top, came down here and was saying there's still some manual steps needed. It was looking for this setup dot m j s file. When I actually went into there, it wasn't in there. I think this is probably now called the Bootstrap gateway file, something like that. And it gives you some recommendations of, like, okay, it's successfully installed. This is what you can do to use that API key that we generated previously up over in here in the app.more.org,

this key. This is how you actually chuck it in. I won't be showing you my key because I don't want you guys using that. But what we then got to was this area over here, where it was just saying like, yep, we've used your bootstrapping. So the install wasn't found, this wasn't found, but the ever Everclaw setup was complete. And so it actually worked relatively simple, relatively easy. You know, little slight hiccup, but the suggested command worked. And so I got to the point where I was then

just kind of using the OpenCLORE setup to actually go or config and using my models and going like, okay, like, what do I actually need to do? Jumping over to here, jumping into the model, and I was able to start seeing it see it down in these areas here. Not sure it actually shows up in this portion at the moment, but I was using Venice just before. And I think there's an option for a,

like, a different type of access to it. You play around with it, you find out what actually works and not. So now we jump into the scariest part of all, which is me attempting a little bit of a a live demo. So we can see I've got my OpenClaw set up here. I've got my current model of Venice, which is at GPT OSS one twenty b. I was using Kimi just before.

And then the Gila GLM 4.7 flash is the gateway. So it kinda prompts you like, hey, what what kind of model do you want to use? And because I was looking at the model pricing, I was going, oh, yep. GLM 4.7 Flash, that's probably the cheapest. And my problem at the moment is because I'm just doing like little bits of testing here and there, is that I just don't have a huge amount of credits. And so I'm always running into, like, oh, I'm running out of credits pretty quickly. So

what we're gonna do is and I'll show you. I'm right at the limit at the moment. I've got 1¢ of credit also using this OpenAI GPT OSS. And so what you can do is actually have a couple of different fallbacks. So I'm just gonna ask it like, hey, chew through some tokens. And what should happen is if I use up this remaining like 1¢ of credit,

it should then default back to one of these other models that I've got up the top here. So in particular, the flash Morpheus gateway. It should default back to that. And then what I'm hoping will happen and so it's creating a fairly long passage through through some tokens. So it's talking about the Megacity of Luminar and just going through that. And what should happen, I guess, is if I asked it another question,

I'm not sure if it'll tell me directly, hey, the model is run out. You run out of credits yet? If it'll tell me directly or if it'll just jump over onto the other GPT sorry. Not GPT. The other Morpheus GAT gateway. Okay. We actually saw it live just flash over there. So I think that actually did it in real time. So that's kinda cool. That's a win for me. I wasn't expecting that to happen. So the yep. And it tells me now the fallback active is on the Morpheus gateway.

And I'm gonna ask it okay. Expand upon that story a little bit more on that story. See if it can do that. And what we should see up the top here is that it'll start using the credits that I had. So if I go over to billing, so it's chewed through my last little credit here, and I would expect it to start to chew so chewed through my last daily API credit remaining

and should now start ticking into this one. So let's jump back to this screen. And wow. Amazing. One more. And let's see if it does it in real time again. Oh, man. That'd be so cool if it actually actually does do it. So let me refresh this. Still at ten. So it could be that I'm just not using up enough or that I'm actually at, like, 0.005

and actually it doesn't show on the dashboard. But in any case, I think that's a pretty good demo of how you can actually use a couple of different providers at the same time and how you can actually use the Morpheus skill to jump in. So, you know, I made it look a little bit easy here because I had done a bit of setup yesterday. But I'll I'll this time, I'm actually really happy because

it was a much less painful process than setting up an actual agent for the first time. So kind of I guess, what I would just recommend is, like I mentioned with those prerequisites, really just working on getting your agent set up first because all of these things are much easier when you can actually talk to your agent and say to it, like, hey, Neo. I want you to switch over or let me know. Is the Morpheus gateway active?

Can you tell me why it's not active? Which model is it using? Can you change that model for me instead of you having to go to the command line like I'm doing here where I am a complete idiot and dumb head and have no idea what I'm doing. You really wanna be getting to the stage where you're just talking to it and it's doing the stuff in the back end. So the morpheusskill.com, that's how you can get access to the API gateway and why you want to use app.more.org

to get your staking key with your API key and using more.org and the builders subnet to be able to do that. And I'll just show here as well that there should be another subnet created which is also on the Arbitrum side of things. I think there was one. In any case, it's it's around there somewhere. Using base is certainly much easier, but you can use

the whatchamacallit? Arbitrum as well. I did see it was possible. Didn't show up just then, but that's very good. So that is it for today. I really hope you got some use out of this that through this tutorial. If you're having any troubles, feel free to hit me up via any of the social links or probably is even easier is just joining the Morpheus Discord and you'll certainly get lots of help within there as well if you're struggling with any of this. If you want $10 of free API credits,

that's the place to go and to try out. And with all of that being said, I do hope you're having a fantastic day wherever you are in the world. Actually, oh, no. Wait. Let me let me bring up the the full screen. I hope you're having a fantastic day wherever you are in the world. Caron out.

Open the transcript file

Hosts and Guests

Hosts

  • Kyrin Down Host Feed-stated

Credit the people on this episode

Be the first to rate this episode

More episodes

All episodes
  1. Video Is Coming! | Morpheus Is About To Get Far More Entertaining

    Nov 19, 202652 minAudio

    Zion, hear me! For Season 2 of the Mere Morpheus podcast I'm going to be uploading the Morpheus builder weekly updates. My reason for doing this is to make it more accessible for those who want an update directly via...

    Credit the people on this episode

  2. Local Models, Foreign Affairs | The Case For Private AI Archives

    Nov 13, 202653 minAudio

    Zion, hear me! For Season 2 of the Mere Morpheus podcast I'm going to be uploading the Morpheus builder weekly updates. My reason for doing this is to make it more accessible for those who want an update directly via...

    Credit the people on this episode

  3. Let's Talk Node NEO | Step By Step Progress

    Nov 6, 202651 minAudio

    Zion, hear me! For Season 2 of the Mere Morpheus podcast I'm going to be uploading the Morpheus builder weekly updates. My reason for doing this is to make it more accessible for those who want an update directly via...

    Credit the people on this episode

  4. Hello From Hermes! | A New Way To Onboard To Morpheus

    Oct 31, 202643 minAudio

    Zion, hear me! For Season 2 of the Mere Morpheus podcast I'm going to be uploading the Morpheus builder weekly updates. My reason for doing this is to make it more accessible for those who want an update directly via...

    Credit the people on this episode

  5. What Agent To Agent Connection Actually Looks Like | Inside The Morpheus Agent Stack

    Oct 23, 202658 minAudio

    Zion, hear me! For Season 2 of the Mere Morpheus podcast I'm going to be uploading the Morpheus builder weekly updates. My reason for doing this is to make it more accessible for those who want an update directly via...

    Credit the people on this episode