trying to be better l Monotheist @labyrinth_alpha
Joined December 2024
- Tweets1K
- Following200
- Followers1.7K
- Likes1.7K
R48 retweeted
Cryptic’s 3 Tips for Trading the Current Market
Cryptic has a highly versatile trading style, but he’s best known for identifying asymmetric small cap opportunities before the wider market catches on, particularly innovative projects experimenting with new ideas.
Some of his most notable early calls include:
Pandora: $1M → $300M
Act: $1M → $400M
Arc: $15M → $600M
Hoodrat: 200k > 20mil
Net: $1M → $70M
Stonk: $1M → $300M
Most recently, he's grown his $10K Pump challenge wallet to $200K in just one month, all public.
Here are his three tips for navigating the current market:
1. Don’t be afraid to go against the crowd
By the time a trade becomes consensus, most of the people likely to buy are already aware of it and already positioned.
Some of the biggest market inefficiencies can be found by looking where the crowd isn’t. That doesn’t mean being contrarian for the sake of it, but don’t be afraid to trust your own instincts and thesis, even if the crowd disagrees.
2. Execute well and size appropriately
Plenty of people can identify a trade that eventually goes up. Far fewer can actually see it through.
Your entry, ability to ride volatility and eventual exit are all part of the trade. Some of the best researchers I know aren’t necessarily the best traders, and execution is often the reason why.
Position sizing is arguably the biggest factor. Oversize and you’re more likely to get shaken out by volatility. Undersize and you may not care enough about the position to give it your full attention.
Get your sizing right and good execution becomes considerably easier.
3. Focus on yourself
The rise of social trading has made this harder than ever. We’re constantly exposed to huge P&L screenshots, making it easy to feel as though we’re underperforming.
That feeling can push you into forced trades, excessive risk and poor execution, all of which will ultimately lower your returns.
Drown out the noise and focus on your own journey. You’re trading for your goals, on your timeline. Don’t let someone else’s results pull you away from your process.
Update: $MLY is backed and built on top of Origin protocol [$40M TVL , $38M raised from tier-1 VCs]
Evidence:
> @PGee13 follows @GoMintly & reposts Mintly's tweets
> @OriginProtocol reposted $MLY's launch day on Virtuals
> Docs, Whitepaper even Website directly mentions Origin protocol as infra has been used to build Mintly
My assumption:
Among all 'utility' launches I've seen so far, Mintly stands out as the most clear, predictable (at least less likely it gets rugged while being supported by gold tier devs), upside-measurable coin.
Being supported by @OriginProtocol which has known as mid/high-tier DeFi protocol (I guess), @GoMintly has more chances for repricing as RH eco evolves and develops.
Rerpicing points:
cashcat gets listed on RH -> some inflow goes to utility coins -> Vex / MLY shares some portion of them
or
Vlad / RH accs mention Virtual -> vex, mly gets pumped
If $VEX got pumped by side mentions, $MLY could play out as more tech aimed coin with known backers & supporters. Also we should admit that RH currently lacks of this kind of projects (Noxa template).
One more thing about $MLY I've forgotten to include is Circulating market cap. that is half of total supply as it goes in docs (app.virtuals.io/virtuals/986…).
nitter.cf/r48_eth/status/2075658…
DeFi + RWA + AI bet on RobinHood.
• RobinHood positions itself as the central hub for the convergence of RWA, DeFi, and AI (including agents).
'An AI-native blockchain built to bring real-world assets onchain.' Source: nitter.cf/RobinhoodCrypto/status…
> At this point, any project building around these verticals will naturally attract far more attention than isolated meta plays.
• Mintly stands as Agentic-first, yield balancer where agent automatically manages user's funds.
• Virtuals have recently given a talk about mintly + @everythingempty is followed the project and personally bought $MLY (as of now, only suggestion).
Source: nitter.cf/GoMintly/status/207523…
> The endorsement makes the setup much more credible and significantly lowers downside and rug risk.
• Unpriced point for Mintly is to let the agents find yields on real assets once RobinHood lists them on chain. I include this point so it's easily implementable on @GoMintly side once it's publicly available.
Mintly has absorbed all the hot narratives that RobinHood has been pushing since mainnet launch.
As shift to utility projects proceeds, market will catch up the plays that requires the initial target of RobinHood chain - the project that builds around DeFi, RWA, AI. Mintly covers all of them.
The direction is clear, the alignment is obvious. Always dyor and nfa, as for me, $MLY stands out as the most assymetric bet on RobinHood + Virtuals ecosystem. Now, let the market decide.
You can copy tech but you can't copy distribution.
Currently, $surplus is aiming to be the biggest protocol by user activity. A few points:
> Current usage is (capital efficiency) as the same level as $RNDR
> Moat strengthens as more frontier models coming and startups/devs are looking for more ways to reduce token costs
> Current valuation [imo] doesn't include the upcoming competition between frontier & open-source models where $surplus stands as the single place to absorb that demand
Surplus serves AI trend as sandbox offering cheap entry to test out any model you'd like to implement into your stack. Regardless, it's for agent, your startup..anything but >70% cheaper compared to frontier subscriptions & LLM aggregators like OpenRouter.
nitter.cf/mac_eth/status/2076276…
DeFi + RWA + AI bet on RobinHood.
• RobinHood positions itself as the central hub for the convergence of RWA, DeFi, and AI (including agents).
'An AI-native blockchain built to bring real-world assets onchain.' Source: nitter.cf/RobinhoodCrypto/status…
> At this point, any project building around these verticals will naturally attract far more attention than isolated meta plays.
• Mintly stands as Agentic-first, yield balancer where agent automatically manages user's funds.
• Virtuals have recently given a talk about mintly + @everythingempty is followed the project and personally bought $MLY (as of now, only suggestion).
Source: nitter.cf/GoMintly/status/207523…
> The endorsement makes the setup much more credible and significantly lowers downside and rug risk.
• Unpriced point for Mintly is to let the agents find yields on real assets once RobinHood lists them on chain. I include this point so it's easily implementable on @GoMintly side once it's publicly available.
Mintly has absorbed all the hot narratives that RobinHood has been pushing since mainnet launch.
As shift to utility projects proceeds, market will catch up the plays that requires the initial target of RobinHood chain - the project that builds around DeFi, RWA, AI. Mintly covers all of them.
The direction is clear, the alignment is obvious. Always dyor and nfa, as for me, $MLY stands out as the most assymetric bet on RobinHood + Virtuals ecosystem. Now, let the market decide.
Robinhood Chain mainnet is live.
An AI-native blockchain built to bring real-world assets onchain.
Learn more: robinhood.com/chain
With $VEX pump, RH ecosystem is about to see its first wave of utility projects backed by @virtuals_io & @bankrbot.
Remember, how chain itself is positioned - AI & RWA, so logical conclusion here is it's +ev to track utility type plays with clean upside.
I did a mistake with $BERRY mostly based on Xen's tweet. On top of that, Berry's dev insolently lied in dms about announcement.
Whatever happens with this project since this day, I won't consider it as legit.
Losses are cut, lessons are learnt.
nitter.cf/r48_eth/status/2074189…
there's a very reasonable point why this setup has become better.
+ Hidden, Forgotten Catalyst by Coinbase coming next weeks.
nitter.cf/r48_eth/status/2068690…
there's a very reasonable point why this setup has become better.
+ Hidden, Forgotten Catalyst by Coinbase coming next weeks.
nitter.cf/r48_eth/status/2068690…
Inference Capital Markets: Considering $JTVO as leading player on Solana.
hot takes:
• @JatevoId is currently processing over 60% of volume on Surplus & UsePod[.]ai which brings 7.5B spent tokens weekly since June.
• Holding $JTVO gives access to higher daily quotas and lower effective cost per inference request. In comparsion with $DIEM, offering value of $JTVO is clearer and more obvious since inference management isn't available for $DIEM holders.
• While AI costs are rising, demand on cheap inference will increase both on company / retail sides. @JatevoId stands out as value recipient on new, growing market.
• At ~$2.4M FDV, $JTVO offers exposure to the hot vertical. Numbers, market coverage, utility, limited direct competitors on Solana in sum makes $JTVO is a single player on this field.
Full thesis:
Main idea of @JatevoId is to capture both frontier & open-source LLM models through offering affordable inference. Think of it like a Market-Maker is trying to fill orderbook in liquidity.
Business model is based on charging users per request. Key point here is the price which can be brought down if user holds a fixed amount of $JTVO. It makes a sensitive difference between $VVV/$DIEM model where Venice users can only use given inference and not able to trade it on p2p markets, this is where $JTVO's model was thought out well.
> Key Point: adding new frontier models will increase demand on $JTVO. Patiently waiting until gpt 5.6 / fable will be integrated.
Besides it, decentralized Inference is one more vertical where @JatevoId is going to step in. Having your own compute to operate open-source models will let team to bring prices down. Same concept, different execution.
Unlock here is to gather idle compute in one place and channel it to providing cheap, affordable models like GLM 5.2, Kimi and others.
In summary, the case for $JTVO rests on proven inference volume, support for both frontier and open-source models, and token utility that adds clean value. With upcoming model additions, it has clear catalysts ahead:
> more frontier models for cheap: Fable, GPT 5.6, Sonnet 5
> Launch of Decentralized Inference network
Seize this setup wisely, especially when market is about to break out of chop market levels. Sources on provided points will be added below.
June Recap.
$pod - 50%
$berry - 2,5x
$dot - 50% [public]
it's been 2 weeks since I came back to trenching and the hardest part of it is upsides. While 2025 base season you were confident about setup and it did work out well on new coins.
At current market conditions, you have to lower your expectations on potential outcome nevertheless token is traded on sol/base.
During this month I discovered that my edge on market is not chasing new launches but sizing into established ones. Execution is, by far, the most difficult part of trading coins to me.
But one thing is and remains clear - supporting real builders and projects always bring net positive outcome to anyone in this space.
At this point, I will push myself forward into finding firm, early, clean setups on the market.
Also, make sure you're surrounded by experienced circle of traders but not candle chasers. Infinitely grateful to @0xUnihax0r & @Labyr1nthhh_ for provided opportunity to prove myself
Mark your mistakes and ensure you become at least 1% better at what you do than yesterday.
More to come.
DotAI: Private CLI to everyone.
@usedotai is about to drop DotCode (currently in beta), you can think of it as Venice but for actual coding.
Why does it matter?
Venice got hyped because it had built a privacy layer between frontier models / interface. So anytime when you interact, you only send the context without getting leaked by your fingerprint , location , previous history of requests and etc
DotCode is the same but for actual coding. Current CLIs have equipped with bunch of limitations that makes difficult to build stuff (see attached screenshot). DotCode fully erased those restrictions and let you to build anything you want
Source: git.usedot.xyz/our-products/… + check demo
What's upside?
DotCode is the first mover in this direction. All restrictions are lifted on the models you're used to use, so the opportunity cost is rising in the biggest audience of devs / users that are busy with vibe-coding.
Same quality, same output, same price in addition to privacy layer makes DotCode is unique. Especially, when it enters in unfilled vertical so upsides are not capped at this point.
What made Venice is great that the entry point was cheap with guaranteed protection of user's data. @usedotai enhances this thesis and brings a tool which lets anyone to stay private while building favorite apps.
At this point, upside on $DOT is clear to me but make sure you've seized your risks and put my thesis under criticism (I may be mistaken as any mutual in this environment).
ALT here's how Codex blocked the entire interaction with agent. It occurs quite often even when you aren't about to hack anyone/anything.
R48 retweeted
You're missing out
t.me/ideazgarden
Tailwinds for base:0xed664536023d8e4b1640c394777d34abaff1df8f.
1. Access to Fable still/remains restricted. The reason is clear - to prevent 'illegal' usage by unwelcome parties. This is the first reason to look at rising DeAI market as a normie.
2. Gpt 5.6 preview on paper has already faced drawback by gov which proved the same argument on frontier-models. As a company, there's no option but follow the order regardless of whether the end user loses access or not.
> Separately, it is worth mentioning that the cutting of limits and the restriction of capabilities in subscriptions to Claude / GPT (an irreversible process) are forcing users to look for alternatives. Local hardware does not deliver sufficient power for the same performance, therefore the DeAI market is *already* experiencing wide demand.
3. Market flush has erased low quality / not convincing setups which is actually good once the market bounces back, plus, I wouldn't exclude that market is always hungry on asymmetric bets have constant non-crypto demand.
In summary, $pod is in perfect shape in terms of the market it is going to capture + the way of how they are going to do it (flywheel, no Vc, no equity, full token dedication).
For the record, v2 is announced but it hasn't come in full form as it has been released. So, make sure market is fully priced in the following updates:
"Post v2 schedule
- GLM 5.2 added
- API live
- Open Router listing
- XMR / SOL / ZEC / BTC accepted as payment for API
- Image gen
- Audio gen
- Venice Uncensored 2 model release (will be partially training this on our network during the RL stage (uses all the nodes inference for the RL rollouts)
- v2 of chat.dphn.ai live with subscriptions"
Source (t.me/c/2607397075/10670)
nitter.cf/dphnAI/status/20704117…
Will be running datagen on the network until we have sufficient capacity - at which point we will open up our inference API to the public
In terms of what we are adding next ->
- GLM 5.2 for 8xH200 / B200 / B300 nodes
- Windows WSL + MacOS support
- Inference API + Open Router
- Audio gen
- Image gen
- AMD GPU + CPU nodes
- Using idle network inference for RL rollouts during model training
- Sharded inference with large models split between many consumer GPUs
(see our initial report on sharded inference over the internet here drive.google.com/file/d/1bzn… )
additional links, good to revisit later:
nitter.cf/iamgingertrash/status/…
nitter.cf/mermachine/status/2070…
TLDR: Still early.
nitter.cf/chjokka/status/2069041…
Base launching b20 is going to be revolutionary, and base:0x778d347b2ffbadf31a2a1be9cf42b4c7ba8b1ba3 is the best way to be exposed to it. here's why:
why is b20 such a big deal?
on-chain stocks today come in two ways, and neither one is real ownership in your hands.
the old way (Ondo, XStocks, all of them) sell you synthetic exposure. a broker sits in the middle as source of truth and you hold a wrapper issued by their permission. you own an IOU, not the equity.
the newer way is actually backed 1:1 with dividend rights, but they're locked to Coinbase itself: permissioned, custodial, no US users.
point is, you never actually hold the real thing,or can do anything with it when holding.
the real unlock is native issuance. an asset that lives on-chain, not a screenshot of one, and not locked behind a gatekeeper. it stops being a thing you just hold and becomes a thing you can use. LP it, collateralize it, route it through a strategy, and more.
Base's b20 is the standard that makes that real.
any issuer can put a compliant asset on-chain at the protocol level, with the toolkit (roles, transfer policies, supply caps, freeze & seize) built into the standard itself instead of bolted on per project.
and the standard part is what people sleep on. b20 is an ERC-20 superset, so every token on it works in existing wallets, dexes and vaults out of the box. which is what turns a few scattered launches into one connected market that can actually scale.
(and that's where @BerryFiBase comes in, building the DeFi layer that sits on top of the standard so every one of those tokens has somewhere to plug into)
so tldr on b20:
RWA on-chain so far was basically stablecoins and tokenized treasuries. real ownable equity barely existed, what's out there was synthetic exposure.
b20 makes native, compliant, permissionless issuance a protocol-level feature on Base, the rail stocks, stablecoins and RWAs can finally be issued on directly.
why is base:0x778d347b2ffbadf31a2a1be9cf42b4c7ba8b1ba3 the coin to buy to be exposed to this b20 protocol launch?
quick on what Berry is: it's the full DeFi / aftermarket layer for Base tokens. LP pools, single-sided liquidity vaults, time-locks, plus a harvest engine that turns idle supply into yield instead of leaving it as dead sell pressure. one thing you can route that yield into is cheap AI inference, through Berry's partnership with Surplus. and the whole stack is already agent-accessible: the juicer is live on bankr skills, so an agent can spin up a pool and earn by command, no manual contract calls.
(they're building this whole DeFi stack out more, as we speak)
and to prepare for b20, Berry is already building a launchpad terminal on Bankr (goes live with the launch of b20 on the 25th of June I believe).
you launch a b20 straight through it and set up the whole token from one place: mint, burn, compliance policies (allowlist, blocklist, freeze & seize), roles and supply caps.
but the post-launch side is where it gets even more interesting.
once your token is live you put it to work in Berry: LP pools, time-locks, earn inference on top of your own b20, and more.
that's the moat. every b20 that launches needs real DeFi wrapped around it, and Berry is one of the few legit teams with that full stack already built / building it out right now.
b20 will be huge for Base, and Berry is building the launchpad terminal + post-launch DeFi standard for all b20 tokens.
I can actually see base:0x778d347b2ffbadf31a2a1be9cf42b4c7ba8b1ba3 become a top 5 coin in the Base ecosystem, and receive a lot of support from Base as they're the only legit team actively building around their new shiny product (b20) that's launching soon.
Revisiting initial thesis for base:0xed664536023d8e4b1640c394777d34abaff1df8f.
They have already achieved PMF (product market fit) by partnership with Venice. Over 46b tokens has been used / burnt (ai tokens). A few reasons why:
> DolphinAI has fine-tuned the most accepted open source model - Qwen36b, fully removed censorship.
> it's cheap. Having held the same quality of model with 30% discount compared to OpenRouter (they charge $1 per 1m tokens, in/output).
Those open-source + uncensored models are getting more and important, especially when leading labs constantly add new layers of restrictions getting pushed by government.
Another thing should be put on the board is the difference between open-source / closed models are getting closer by the quality of output. Current achievements include reaching Opus 4.6 / gpt 5 model levels which is insane.
base:0xed664536023d8e4b1640c394777d34abaff1df8f wins here even more, after all, the Decentralized Network everyone is hyping about will be delivered in upcoming infra update.
So the frame or flywheel looks like this:
Claude/Gpt model release -> Chinese quants exctract thinking / output , prepare open-sourced model [qwen, minimax, GLM, Kimi] -> Dolphin fine tunes the model, removes all boundaries -> list the model on Venice , OpenRouter -> anyone can process users' prompts and earn $ on idle gpu -> ppl love it more because you're able to use model without any restrictions -> base:0xed664536023d8e4b1640c394777d34abaff1df8f pays GPU suppliers + buyback the token -> repeat
I would include one more point - Decentralized Network will let you deploy listed models you wish and use it securely, privately with full removed boundaries. I don't think this have been priced in on token side yet.
plus, catalysts which will gas the repricing
> API launch
> v2 launch
> Decentralized network release (this is what $zero trying to achieve, already will have been served by Pod)
> OpenRouter listing
all comes starting from Wednesday
Don't forget to send some credits to @jkrdoc, he's gathered all these arguments in depth written article (pod-research.netlify.app/), bullish.
nitter.cf/r48_eth/status/2067932…
A few thoughts on $POD.
I've been trenching for ~2 weeks and can say it for sure that we have several clear meta that normies/retail are about to accept broadly. It is:
> Decentralized AI
We've already crossed the line where AI enjoyers moved back to private inference and been using frontier models while queries are encrypted.
But it's a temporary solution to this problem since frontier models are still limited and censored. Open-source models have solved this issue, however, most just don't have enough capacities to maintain the infrastructure. RAM + energy issues.
That's why we need DeAI. The idea itself is already evolving - numerous GPUs working together on processing prompts sent by users, not knowing the author of initial request (it's called 'private inference').
@dphnAI does exactly this. Distributed network of GPU suppliers are being paid in $POD for processing requests through interface + APIs.
What makes Dolphin different is flywheel tied to the token. The idea of decentralized + private computing can't be brought by VCs subsidized project. This angle makes $POD way flexible in the context of providing
private, secure infrastructure than competitors (OpenGradient / OpenRouter).
My thesis for $POD is further distribution + reprice based on open-source models' development. The better models become the more @dphnAI's part of pie grows.
Cheap + private + same quality (compared to gpt/claude) inference will play self-fulfilling short term thesis against frontier models.