been a fan of @luna_virtuals and @virtuals_io early on
an underappreciated aspect of Luna is the underlying flywheel effect
-> Luna captures the upside of Virtuals (as its token is paired & coupled w it)
-> captures the upside of Luna itself
what’s the upside? why Luna?
@Fiskantes you are talking about @virtuals_io which platforms all of the above
e.g. their ai agent/influencer LUNA has 500k+ tiktok followers, more than 10m+ likes, does non-stop livestreams on youtube and tiktok
and has its own $LUNA ticker (now at 4m mcap), which she now shills
We are excited to announce that @printr is now officially live and integrated on our trading terminal! $PRINT
You are now able to:
- Trade Printr tokens with instant execution
- Monitor, track & identify trending tokens early on
- Capture 20% of all the revenue generated by the terminal
@0xNairolf yeah, happy to see it because it just makes sense
app layer is where all the user ownership is
and hence where most of the value should accrue
You know what, I'll match Ross, Dan and Mert here.
If you drive $1M in monthly volume through our terminal to ANY asset
I'll write you a 20% profit-share check, give you pitch deck review advice, and provide as many complimentary intros as you'd like.
Bonus if you use @solonamobile apps, a @dflow API and a @Helius plan.
@0xNairolf yeah, essentially everything will become an aggregated experience
kind of fascinating how clear that is, yet how few build around that
i now have a mandate to only invest and back founders & initiatives that are conscious and proactive about it
@dcfgod yh was super turbulent and unstable all around
from all normal -> banned -> switched wifi provider -> UI re-appearead -> watched contract -> sudden 40m fill instantly -> deposited -> filled on all size
@ashleighschap I'd argue Uniswap's protocol design speaks for itself.
Her take reflects someone who never actually touched or *understood* the technical side, which shapes the way she reads the entire space..
It fits the perspective you usually get after spending years inside the SEC bubble
Holy shit... Tencent researchers just killed fine-tuning AND reinforcement learning in one shot 😳
They call it Training-Free GRPO (Group Relative Policy Optimization).
Instead of updating weights, the model literally learns from 'its own experiences' like an evolving memory that refines how it thinks without ever touching parameters.
Here’s what’s wild:
- No fine-tuning. No gradients.
- Uses only 100 examples.
- Outperforms $10,000+ RL setups.
- Total cost? $18.
It introspects its own rollouts, extracts what worked, and stores that as “semantic advantage” a natural language form of reinforcement.
LLMs are basically teaching themselves 'how' to think, not just 'what' to output.
This could make traditional RL and fine-tuning obsolete.
We’re entering the “training-free” era of AI optimization.
@delucinator hey, im not too active on here but have been reading ur tweets for a while
if u got carried out and need some cash to get back, lmk. happy to help
@dcfgod $2.193B TVL
$157.48M annualized fees
$11.67M annualized revs
$126M fdv
and a portion of fees now approved to be redirected to flow back to the token?
only $MNDE
@0xGeeGee curious why?
you think that the only tokenized exposure to prediction markets + perps + socialfi has a ceiling at 20m fdv?
feels like its only getting started and the comparables are valued an order of magnitude higher