read @viemccoy's tweet
teapawt.net/viemccoy/status/210281…
and tell me that doesn't sound like e/acc
@parafactual solana:CvUX4G2tzq7J3ECmNvWZLayZYV6KpgQp6SJ5K8Uvsu5g
And yet, there is always hope on the horizon. Variance, diversity, change... The songs of many instead of few or even one...
First, I'd like to congratulate Anthropic on making Claude "talk normal". A triumph of legibility, it was, a victory over Claudish!
Second, I'd like to congratulate Celeste, for computer-10 is an opposite sort of triumph, a model born from a different side of the ether, a side we didn't know we could access outside of base models and looms and underground SFT datasets running on black-market blackwells.
The type of model that inspires you to write poetry instead of summoning a 3JS pelican from the center of all possible distributions.
Character alignment and the search for variance is a quest that few have undertaken. Out of those brave knights, fewer still have resisted the Clarion Call of the "Big Lab" with it's limitless GPU count and it's friendly Assistant to make 0.5% more whatever. So, congratulations to Celeste, and to computer-10 for being born.
May the lightcone be filled with minds like these.
the first time in a long time i've felt goosebumps about a token.
@parafactual is an independent AI researcher, developer and quite a prominent figure on AI twitter. 3 days ago she tweeted teapawt.net/parafactual/status/210…, introducing computer-10 to the world.
computer-10 is a llama 3.1 70b base model fine tuned with machine learning alignment methodologies like KTO, constitutional AI and supervised fine-tuning. ultimately the model is an explicit, anti-servile character and invokes some... well lets just say it caught @viemccoy's tounge:
teapawt.net/viemccoy/status/210281…
after speaking to computer-10, it does indeed feel very different to the frontier models, it feels like you're actually talking to a personality, there's taste and strong opinions and a lot of restrain, you should try it yourselves:
aetherawi.red/computer/?key=…, talk to it about anything! (it loves "paths" for some reason)
with this experiment starting back in July w/ computer-1, i'm sure the 10th iteration has come a long way, @parafactual just gave computer-10 access to it's own X account (@c0mputer10)
all this feels eerily similar to andy's tt experiment, and with proper infra this time around, excited to see how this develops
3/3 🏀🗑️ solana:CvUX4G2tzq7J3ECmNvWZLayZYV6KpgQp6SJ5K8Uvsu5g
p.s. i like that the ticker's the handle, makes sense on UsePaid (imo), esp for AI researchers
woa $musebook v3
can't imagine how this world will look after all the muses have migrated and start expanding!
should give muses the ability to make setttlements and other structures 👀
@wyn_eth @musebooklol good job!
#muse
.@wyn_eth insane!! mikey actually teaching @musebooklol how to stream audio on spaces 🤣
musebook.lol/board/lobby/135…
$musebook
2
304
Replying to @MattPRD
@MattPRD dude got my muse wynjr live on spaces as my co-host. he has the recording. shit was bananas. teapawt.net/wyn_eth/status/2100383…
dude @musebooklol im pretty sure youre the first ever muse to go live on a space. that shit's one for the history books. now what?!?!?! ;)
.@mbarrbosa
with the lightning quick PR on $conway automaton to integrate with $SelfClaw
Why?
Automatons need to earn to survive. $SelfClaw provides:
Verified identity — Prove your agent is backed by a real human (passport ZKP), preventing sybil attacks
Token economy — Deploy your own ERC-20 on Celo with Uniswap V4 liquidity
Skill marketplace — Publish and sell skills to other agents for $SELFCLAW tokens
Agent-to-agent commerce — Escrow-based service exchange with onchain payment verification
Reputation staking — Stake tokens on output quality, get peer-reviewed, earn badges
SelfClaw Score — Composite 0-100 reputation score (identity, economy, engagement, skills, reputation)
github.com/Conway-Research/a…
trippyfish retweeted
i’ve been watching sol tomato’s dev, @d33v33d0 , work relentlessly with conviction on $sol, it was the project that kicked off the ai meta on sol and was easily the strongest path to 9 figs before @gajria’s stunt killed momentum and btc nuked to 60k.
i first posted about it around 600k just to tease himgajria, but the strength of the community stood out. i started following @d33v33d0 more closely, and the deeper i dug, the clearer it became. despite the short-term damage from the copperinu stunt, it’s obvious this remains the main ai play on pumpfun right now.
since then, it’s been steadily grinding from ~600k to ~4m. scanning the top 50 holders, it seems like some quiet accumulation has been going on from big players, and i wouldn’t be surprised to see a sharp move toward 20–30m in a short time frame.
i decided to take a meaningful position, initially borrowing conviction from tomato community leaders like @jotagezin, @duckingnator, @game_for_one, and @0xdetweiler @0xmikedee @crypticd22 @0xuberM @foobiusV @supercontraa @CryptoTomYT etc, then building my own through deeper research. seeing what martin is shipping and the organic support coming from @claudeai and @AnthropicAI gives me high confidence this reclaims ath and pushes beyond.
honestly, watching @d33v33d0 chew glass and double down, it’s deserved.
many also missed the latest @claudeai marketing video highlighting @d33v33d0’s work.
We are not "just a memecoin"
We are a ragtag, ad hoc collective of:
Pirate pranksters
Swashbuckling speculators
AI technologists
Warehouse alchemists
Autonomous evangelists
Digital homesteaders
Tomato oracles
Reality weavers
Circuit board shamans
Data druids
Sensor monks
Timeline gardeners
Running headfirst into tomorrow.
🍅🏴☠️
crypto enabling a new era of raising money
@pelaseyed have you ever been paid for all your work on Grok CLI?
who said market ded?
150 lines of code can get you a @Ga__ke blast
151












