🚀 1,000+ TOKENS/S ON A 1T MODEL! 🚀
We are thrilled to release Xiaomi MiMo-V2.5-Pro-UltraSpeed in collaboration with @TileRT_AI , breaking the 1,000 tokens/s output speed on a 1 Trillion parameter model for the FIRST TIME!
Not wafer-scale integration like Cerebras. Not pure on-chip SRAM chips like Groq. We achieve 1,000 tps on a 1T MoE model using just a SINGLE, STANDARD 8-GPGPU NODE.
Read the full technical deep dive:https://t.co/MX0kjHKdKi
Want to experience the future of real-time AI?
👉 Apply for UltraSpeed now: https://t.co/aeWAxyhwVk
⏳ Limited-Time Access: Application-based · Jun 8 – Jun 23 (PDT)
💬 Chat Experience: Completely FREE for a limited time — try the blazing-fast web chat now.
⚡ UltraSpeed API: Just 3x the price for a ~10x boost in output experience.
🤝 Enterprise & Large-Scale Needs: [email protected]
Whenever I don’t use codex for a task, I ask myself why and usually realize that there’s some missing context, I needed to write a skill, or I just didn’t think to use it.
Rarely is it because the task is outside of the capabilities of the model. Overhang right now feels large.
Building AI that Builds AI: Introducing the Sakana AI RSI Lab 🚀
https://t.co/AskX3J5oEJ
Today, we are announcing the Sakana AI Recursive Self-Improvement (RSI) Lab: a dedicated research group in Tokyo tasked with redesigning the AI development process itself using AI.
While the industry increasingly speculates about the theoretical potential of self-improving AI, we’ve spent the last two years actively laying the foundations to make it a reality:
▪ LLM²: AI models automating research to invent better preference optimization algorithms.
▪ Darwin Gödel Machine: Agents autonomously rewriting their own codebase to double software-engineering performance.
▪ ShinkaEvolve: Hyper-sample-efficient program evolution that builds novel loss functions for MoE models.
▪ ALE-Agent: Reinforcement agents outperforming hundreds of human experts via self-learning.
▪ Digital Red Queen: Open-ended adversarial coevolution laying the groundwork for RSI in cybersecurity.
▪ The AI Scientist: Towards end-to-end automation of AI research, recently published in Nature.
Now, we are unifying these breakthroughs. The Sakana AI RSI Lab is officially tasked with building open-ended, adaptive architectures that collectively self-improve.
Human intelligence did not emerge from limitless resources; it was forged through the open-ended, compounding process of evolution operating under strict constraints. We are applying this exact principle to AI.
We believe recursive self-improvement is achievable on modest, sample-efficient compute. It shouldn’t be a winner-take-all asset locked inside hyperscale clusters, but a democratized public good.
We’re scaling our team to execute this mission. We are looking for frontier scientists and engineers who are entirely unsatisfied with the brute-force status quo. If you are ready to break away from standard benchmarking and build the self-improving future in Japan, come build with us.
30-40K is giving me the same vibes as the 10-12K calls last cycle.
It became the consensus view, but it never happened.
So forgive me if I'm 10% early. I was about 10% early calling the top at 112K $BTC, and that still worked out in the end.
A lot of what I do today comes from mistakes I've made in the past. Mistakes that cost me a lot of money, mistakes I have no intention of repeating.
A lot of people criticize my trading style. They criticize the gradual TWAP entries. They criticize what they see as an "unreasonable" stop loss. The reality is that most of them simply don't understand the mathematics behind how I trade.
I use low leverage. I position myself accordingly and calculate the statistical probability of certain price levels being hit, using leverage as my invalidation point. I'm not a gambler, I'm a compounder.
I'm here to make money. I don't care about quick gains, and I don't care about the reckless trading people engage in for clicks and attention. I try to position myself the same way institutions and hedge funds do.
Slowly, gradually, with low leverage, and only when the RR is significantly in my favour.
I think a lot of people underestimate how much wealth can be built through compounding over a long period of time because they're constantly chasing fast gains.
The reality is that wealth is built over years, not weeks.
The longer you stay in the game, the more disciplined and patient you are, the more money tends to follow.
It's easy to develop a distorted perception of trading with the amount of crap you see on this app. People posting outrageous gains, bragging, and chasing engagement.
But they'll never show you the losses.
If that temporary ego boost gives you satisfaction, great. For me, not so much.
So, simply put, I'm here to make money, whether it's fast or slow.
If you don't like my passive approach, I couldn't care less. This is how the biggest players in the world operate. You can either position alongside them or against them.
I choose to play with them.
SpaceX was founded to make life multiplanetary. We’ve been able to expand that mission with our Starlink constellation and AI solution
Learn more → https://t.co/PSCyWrMUYI
one of the quotes i find most inspiring on a hard day:
"Whatever your hand finds to do, do it with all your might, for in the realm of the dead, where you are going, there is neither working nor planning nor knowledge nor wisdom"
Ecclesiastes 9:10