vallettabeats retweetledi
vallettabeats
2.8K posts
vallettabeats
@vallettabeats
@Twitch Content Creator | @MilkersGGs | business inquiries: [email protected] 🇲🇹🇵🇸
Katılım Temmuz 2016
585 Takip Edilen341 Takipçiler
vallettabeats retweetledi
Weekly Community Update
Over the past week, it’s been continuous grinding towards the next phase of the project.
We didn’t expect this to take this much time, or maybe we underestimated it a bit. But mainly, the testing phase is taking longer than expected. We’re making sure everything actually works the way it should at scale.
And now, things are starting to come together.
> AlphaStriker Update
Tomorrow, we roll out the next version of AlphaStriker to private beta testers.
All the feedback from the community has been implemented with Major UI improvements and better execution with multiple modes easily one of the best experiences you’ll have in a trading bot.
It’s smooth, addictive, and effortless to use.
and we’re confident you all will love it on public release.
> Infrastructure Upgrade
The backend system has gone through a full upgrade.
Alerts are becoming more contextual, smarter, and deeply integrated with social intelligence.
New alert types are coming.
and the infrastructure is built to handle serious scale with millisecond-level execution.
Development is complete. Currently in testing.
Rollout expected next week.
> Social Engine
The social engine is ready.
We are holding the release until the information layer upgrade is live so you can actually use it at full potential.
This is not just a product.
This is a full infrastructure being built piece by piece, where everything connects and works together.
We’re not rushing it.
We’re building it right.
Big weeks ahead.
English
vallettabeats retweetledi
WEEKLY COMMUNITY UPDATE
this week was all about getting ready
ready for what?
expansion and distribution
we’ve been deep in the trenches building the backbone for what’s coming next
> optimised latency and upgraded infrastructure to handle thousands of users smoothly with near instant response
> laid a stronger foundation to support upcoming features without friction
> improved UI for personalised alerts setup and management
> refined the telegram trading bot interface for better execution and usability
we underestimated how deep this phase would go, but it pushed us to build things the right way
the last 2–3 weeks may have felt slow from the outside, but internally it’s been heavy, focused work
we’re ready to move louder soon
English
$iroha just hit $2M mcap and you probably saw it too late
it’s a japanese baby hippo and now everyone’s talking about how cute it is and comparing it to moodeng
but where was that information before the move?
and if you had the right info at the right time what were the chances you would’ve actually caught it early and many other runners early
but most retail traders simply don’t have that edge
and it’s not even your fault that’s just the current state of retail chasing pumps reacting to noise relying on influencer calls by the time it hits your timeline and trending sections it’s already too late and priced in
and even if you get lucky and enter early it gets hard to hold because there is no conviction and conviction comes from the right information and understanding narratives
so we built something to close that gap completely for retail
yesterday something really interesting happened
while testing our ai agent the agent directly shilled $iroha at $50k mcap when it saw the social narratives and smart money movements aligning properly it didn’t even present the info it shilled it
look how good it is in
> understanding the narratives and origin
> understanding the virality factors and quality of narrative
> on chain data and smart money movements
> deep sentiment analysis
> deep understanding of human psychology and behaviour
understanding and connecting everything properly together and generating insights never seen before
the best part
it was completely autonomous no human intervention involved at any point picking up coin from our on chain engine to pulling its social data analysing and processing everything and posting on X when no one was talking about it
and the move turned into a clean 40x return
and the funniest part is the coin was literally sitting at $50k for hours before the move clean entry window if you had the right data at the right time and conviction to hold which only comes from proper information
this is just a preview of what’s coming
we are still refining the agent and tuning it but this was one of those moments that clearly shows what the system is capable of
below is the exact message from the agent read it carefully level up your game and be ready for what’s coming next 👀
English
@BobbyBigSize @AlphaBlockAI That’s Bobby Diamond Hands right there baby 💎
English
@AlphaBlockAI if you’re still sleeping on $ALPHA after reading this idk what to tell you
if this doesn’t wake you up nothing will
English
@AlphaBlockAI RIP to all the non-believers selling bags for pennys, we are so primed.
English
Weekly Community Updates
> Over the past week AlphaStriker our Telegram trading bot has been one of our main priorities and the core infrastructure is now complete with a strong foundation in place. This is not going to be a basic trading bot as from day one it is being built to handle complex trading situations including proper position management, TP/SL, limit orders, fast execution, low latency and optimal trade fills while also designed to work seamlessly with your personalized information layer
The infrastructure is now stable and reliable and was built with the help of our community through real time feedback during closed beta testing. Our early users were not just testers they were co builders in this process and helped shape the product with us. With the core infrastructure now complete we will begin a rigorous testing phase this coming week with our private beta testers to make the system fully battle tested before the public release
> We also want to acknowledge that especially this phase of development took longer than we initially expected but building strong foundations takes time and we believe this is part of the process We are excited to share that alongside the Telegram bot launch we will also be rolling out our Social Intelligence Layer completing three major pillars of AlphaBlock AI
• Social intelligence layer
• Telegram execution layer
• A new and improved personalized information layer connecting everything together
> Regarding our organization verification badge on X (golden checkmark) it was temporarily removed due to what was described as suspicious payment activity We believe this is simply a misunderstanding and according to X policies and our compliance everything is in order. We are already in contact with the X support team and expect the badge to be restored soon so there is no reason for concern
From the beginning our goal has been clear to build a system that combines onchain data and social data processed through AI and delivers it in a personalized way that helps you make independent and better informed decisions instead of relying on others everything is being built to put you in control and we are now closer to that vision than ever before
This was our building phase
Now we are about to enter the expansion phase
And we are extremely excited about what comes next
with appreciation
Team AlphaBlock AI
English
From now on these community updates will drop every Sunday
I want this to be a consistent check in so you always know where the project stands what we’re building what’s coming next and even the problems we’re facing along the way with full transparency
Planning to bring these into video format soon as well to make them more insight rich more engaging and easier to follow with some cool breakdowns
Transparency builds trust and trust builds communities that actually last
The best updates are still ahead
See you every Sunday
Alpha AI@AlphaBlockAI
Weekly Community Updates > Over the past week AlphaStriker our Telegram trading bot has been one of our main priorities and the core infrastructure is now complete with a strong foundation in place. This is not going to be a basic trading bot as from day one it is being built to handle complex trading situations including proper position management, TP/SL, limit orders, fast execution, low latency and optimal trade fills while also designed to work seamlessly with your personalized information layer The infrastructure is now stable and reliable and was built with the help of our community through real time feedback during closed beta testing. Our early users were not just testers they were co builders in this process and helped shape the product with us. With the core infrastructure now complete we will begin a rigorous testing phase this coming week with our private beta testers to make the system fully battle tested before the public release > We also want to acknowledge that especially this phase of development took longer than we initially expected but building strong foundations takes time and we believe this is part of the process We are excited to share that alongside the Telegram bot launch we will also be rolling out our Social Intelligence Layer completing three major pillars of AlphaBlock AI • Social intelligence layer • Telegram execution layer • A new and improved personalized information layer connecting everything together > Regarding our organization verification badge on X (golden checkmark) it was temporarily removed due to what was described as suspicious payment activity We believe this is simply a misunderstanding and according to X policies and our compliance everything is in order. We are already in contact with the X support team and expect the badge to be restored soon so there is no reason for concern From the beginning our goal has been clear to build a system that combines onchain data and social data processed through AI and delivers it in a personalized way that helps you make independent and better informed decisions instead of relying on others everything is being built to put you in control and we are now closer to that vision than ever before This was our building phase Now we are about to enter the expansion phase And we are extremely excited about what comes next with appreciation Team AlphaBlock AI
English
@BashKOsman Take profit and add some back on pullbacks. Dips will more often go lower than expected after new highs are created. Don’t chase the top. If there’s real conviction in the coin (@AlphaBlockAI being a great example!), secure more tokens at a lower price for the hold long term.
English
@memesinthewild @anoncoinit someone still knows something with these buys coming through today 👀
English
Got hacked last summer so i walked into a store and bought a whole new PC with zero knowledge of hardware.
Months later I started running AI locally on my machine and discovered I accidentally own an RTX 5090, literally the most powerful consumer GPU on the planet.
I didn't choose the hardware life. The hardware life chose me. 🤣
2 months ago I couldn't tell you what VRAM was.
Now I'm arguing about DDR5 speeds, model quantization, and optimal RAM configs for running 70B parameter models locally.
The local AI wave will turn everyone into a hardware nerd.
You've been warned!
English
@sudoingX What would be a good model to run on 3080ti12gb?
English
spent the entire day testing Qwopus (Claude 4.6 Opus distilled into Qwen 3.5 27B) on a single RTX 3090 through Claude Code. this is my new favourite to host locally.
no jinja crashes. thinking mode works natively. 29-35 tok/s. 16.5 GB. the harness matches the distillation source and you can feel it. the model doesn't fight the agent.
my flags:
llama-server -m Qwopus-27B-Q4_K_M.gguf -ngl 99 -c 262144 -np 1 -fa on --cache-type-k q4_0 --cache-type-v q4_0
if you want raw speed, base Qwen 3.5 MoE still wins at 112 tok/s. but for autonomous coding where the model needs to think, wait for tool outputs, and selfcorrect without stalling, Qwopus on Claude Code is the cleanest setup i've found on this card.
i want to see what everyone else is running. drop your GPU, model, harness, flags, and tok/s below.
doesn't matter if it's a 3060 or a 4090, nvidia or amd. configs help everyone. let's push these cards to their ceilings. let's make this thread the reference.
Sudo su@sudoingX
Qwopus on a single RTX 3090. Claude Opus 4.6 reasoning distilled into Qwen 3.5 27B dense, running through Claude's own coding agent (claude code). 29-35 tok/s with thinking mode on. the jinja bug that kills thinking on base Qwen doesn't carry over. harness and model matched. the base model would pause mid task on Claude Code. just stop generating. that's why i ran it through OpenCode, which handles stalled states automatically. this distilled version doesn't stall. it waits for tool outputs, reads them, selfcorrects when something breaks, and keeps going. i gave it a benchmark analysis task. went 9 minutes autonomous. wrote a README nobody asked for. zero steering. video is 5x speed but fully uncut. if you have a 3090, you can run this right now. free. no API. no subscription. opus structured reasoning on localhost. octopus invaders is next. same prompt that base qwen passed in 13 minutes and hermes 4.3 failed on 2x the hardware. i want to see if the distillation changes the outcome or just the style. more data soon.
English
vallettabeats retweetledi
Some of the most successful coins were CTO'd. But the current process of CTOing a coin has many flaws and results in unfair decisions which hampers the community and its growth.
Our thesis with CTO is simple - give the power to the community to make a decision. We provide the platform to make it happen.
We are now in the final phase of introducing the CTO feature in the platform. We appreciate your patience and hope this would be a new way for the community to come together and coordinate.
English
16k buy just now on $ALPHA ifkyk
3wtGGWZ8wLWW6BqtC5rxmkHdqvM62atUcGTH4cw3pump @AlphaBlockAI @AlphaWhalesX stop sleeping this is the real alpha. ~1m mkt cap is a stealll
English