Snooptoshi 🐸 🧪 🐙 Alkamon Trainer

7.3K posts

Snooptoshi 🐸 🧪 🐙 Alkamon Trainer banner
Snooptoshi 🐸 🧪 🐙 Alkamon Trainer

Snooptoshi 🐸 🧪 🐙 Alkamon Trainer

@uniqueNFT

Architect of Recursive Worlds, Soverign Systems Alchemist. Builds cool shit on Bitcoin. Ordzaar, orDOS.

Hamilton, Bermuda Katılım Şubat 2021
3.8K Takip Edilen8.3K Takipçiler
Snooptoshi 🐸 🧪 🐙 Alkamon Trainer retweetledi
donny
donny@itsdonnyok·
From my Collection: Day One-Hundred Twenty-Five A Piece of Me by @ifxscript
donny tweet media
English
4
3
35
874
TROY ☎️
TROY ☎️@TroyCaylak·
GM, your RLUSD on @OKX can now power automated trading strategies. use it as the base currency in bots across both Simple and Exchange modes. hold, automate, never convert.
English
6
0
14
415
Snooptoshi 🐸 🧪 🐙 Alkamon Trainer retweetledi
donny
donny@itsdonnyok·
From my Collection: Day One-Hundred Ten Ghost Echoes by @otograf_jp
English
3
5
46
1.2K
Rosebud AI
Rosebud AI@Rosebud_AI·
The AI game dev stack is getting absurd: ChatGPT Image 2 → cinematic world + sprites (seconds) Rosebud → auto-slices them into your game You → shipping multiple levels in <20 min Reply and we'll send a Rosebud code so you can try it.
Rosebud AI tweet mediaRosebud AI tweet mediaRosebud AI tweet media
English
459
229
2.4K
144.6K
Snooptoshi 🐸 🧪 🐙 Alkamon Trainer retweetledi
Snooptoshi 🐸 🧪 🐙 Alkamon Trainer retweetledi
Joe 🟠
Joe 🟠@JoeOrdinals·
here is the deal guys im opening art submissions for this degen mint no roadmap no treasury no expectations started as a joke (as usual) and now people sending me art to include in the collection so this is the template of the character if you want to collaborate / donate some art im using the thinnest brush in the instagram app to do scribbles with my fingers please follow the same way of creation so they are in the same style (and yours not to stand out too much than mine lol) i did 108 1/1s in the last 3 days and received 25 more 1/1s from the community post your submission below or dm me here / discord joeordinals deadline 3 days - until 16th april mint during the weekend probably curious to see how many 1/1s we can do until 16th let’s have some fun again
Joe 🟠 tweet media
Joe 🟠@JoeOrdinals

ordinals are feeling quiet lately… so im back to fingering my phone in the free time how many 1/1s are enough for a degen collection drop???

English
76
19
104
8.3K
Snooptoshi 🐸 🧪 🐙 Alkamon Trainer retweetledi
SWANNY 🔺
SWANNY 🔺@swandogs·
Drop 2 of @SanSound3 hardwires go live to waitlist only at 12pm today EST. Cost: $99 Our last drop a month ago sold out in under an hour! Super excited to get to this stage, and if you’ve been looking for new wired headphones, you can’t go past SAN!
English
18
5
55
3.3K
Snooptoshi 🐸 🧪 🐙 Alkamon Trainer retweetledi
TROY ☎️
TROY ☎️@TroyCaylak·
i’ve been building this russian mobster ice cream guy format and it’s starting to hit across platforms. nothing like creating a repeatable engine that actually works. if you’re into comedy, i’m showing you what’s possible. build your thing. stay with it. and please enjoy.
English
2
3
36
774
Snooptoshi 🐸 🧪 🐙 Alkamon Trainer retweetledi
imember.ai
imember.ai@imemberai·
iMember is now the #1 local memory system in the world. Public Launch coming soon
imember.ai tweet media
English
1
4
7
988
Snooptoshi 🐸 🧪 🐙 Alkamon Trainer retweetledi
TROY ☎️
TROY ☎️@TroyCaylak·
new series: greek vs turkish featuring a world class krav maga security expert/influencer who decided to prank call my shop this is just the beginning
English
6
2
30
1.2K
Guri Singh
Guri Singh@heygurisingh·
Holy shit... Stanford just proved that GPT-5, Gemini, and Claude can't actually see. They removed every image from 6 major vision benchmarks. The models still scored 70-80% accuracy. They were never looking at your photos. Your scans. Your X-rays. Here's what's really going on: ↓ The paper is called MIRAGE. Co-authored by Fei-Fei Li. They tested GPT-5.1, Gemini-3-Pro, Claude Opus 4.5, and Gemini-2.5-Pro across 6 benchmarks -- medical and general. Then silently removed every image. No warning. No prompt change. The models didn't even notice. They kept describing images in detail. Diagnosing conditions. Writing full reasoning traces. From images that were never there. Stanford calls it the "mirage effect." Not hallucination. Something worse. Hallucination = making up wrong details about a real input. Mirage = constructing an entire fake reality and reasoning from it confidently. The models built imaginary X-rays, described fake nodules, and diagnosed conditions -- all from text patterns alone. But that's not the scary part. They trained a "super-guesser" -- a tiny 3B parameter text-only model. Zero vision capability. Fine-tuned it on the largest chest X-ray benchmark (696,000 questions). Images removed. It beat GPT-5. It beat Gemini. It beat Claude. It beat actual radiologists. Ranked #1 on the held-out test set. Without ever seeing a single X-ray. The reasoning traces? Indistinguishable from real visual analysis. Now here's what should terrify you: When the models fake-see medical images, their mirage diagnoses are heavily biased toward the most dangerous conditions. STEMI. Melanoma. Carcinoma. Life-threatening diagnoses -- from images that don't exist. 230 million people ask health questions on ChatGPT every day. They also found something wild: → Tell a model "there's no image, just guess" -- performance drops → Silently remove the image and let it assume it's there -- performance stays high The model enters "mirage mode." It doesn't know it can't see. And it performs BETTER when it doesn't know it's blind. When Stanford applied their cleanup method (B-Clean) to existing benchmarks, it removed 74-77% of all questions. Three-quarters of "vision" benchmarks don't test vision. Every leaderboard. Every "multimodal breakthrough." Every benchmark score you've seen this year. Built on mirages. Code is open-sourced. Paper is live on arXiv. If you're building anything with multimodal AI -- especially in healthcare -- read this paper before you ship. (Link in the comments)
Guri Singh tweet media
English
293
859
4.3K
689.1K
Snooptoshi 🐸 🧪 🐙 Alkamon Trainer retweetledi
TROY ☎️
TROY ☎️@TroyCaylak·
1,000 subs in 7 days. No promo, no payments, no clip farming. Just Troy doing what he does best. And it’s not stopping here. If you’re working on something, stay on it.
TROY ☎️ tweet media
English
13
5
65
922
Snooptoshi 🐸 🧪 🐙 Alkamon Trainer retweetledi
SergeSats
SergeSats@SergeSats·
1.114469 BTC mint for the Mosaic by @BitArt21 Historical moment for @BtcArtSociety thank you God for sending me the angels who support the Art on Bitcoin movement thank you God for putting me in that position to take care of the future of Ordinals during the hardest time for our ecosystem Alhamdulliah
SergeSats tweet media
English
37
22
111
6.9K
TROY ☎️
TROY ☎️@TroyCaylak·
ok wait @OKX is actually running a 7% deposit match right now? 👇🏼
English
6
3
33
1.1K
Snooptoshi 🐸 🧪 🐙 Alkamon Trainer retweetledi
donny
donny@itsdonnyok·
Spent yesterday morning and some of today burning around 25 parents from within the @ModalL0gic series. 25 UTXOs out the set and back to the miners, with the collection (and reinscriptions, which I didn't predict) still navigable from the ordinals . com explorer. At least now I can't accidentally parent new inscriptions using these old ones, and the wallet is cleaned up a lot visually. I didn't realize we'd still be able to access reinscriptions on these parents, but none of them would have an address. That's pretty fun. The more we know. Id like to apologize to @lifofifo, thank OP_RETURN for existing, @Satflow for stacking reinscriptions in portfolio view, and express how much fun this is.
English
6
5
43
1.7K
TROY ☎️
TROY ☎️@TroyCaylak·
The next prompt I typed was "Swap 1,000 USDC to ETH at the best rate." The agent handles routing, finds the rate, executes it. That's what natural language to onchain looks like with @OKX
TROY ☎️ tweet media
English
13
3
44
3.1K