Building self-improving superintelligence CEO @recursive_si and @youdotcom MP @aixventuresHQ Ex: Stanford Adj Prof, Chief Scientist at Salesforce, CEO MetaMind
Joined May 2013
- Tweets5.7K
- Following1.2K
- Followers128K
- Likes20.7K
Pinned Tweet
Introducing Flint, @Recursive_SI’s first model: a fast local coding model built with our automated research system, the Eureka Machine.
• Best-in-class fast local coding
• Up to 2x faster than comparable models: 125+ tokens/sec on a MacBook Pro, 150+ on RTX Spark
• Built from scratch in under 4 months
Private, unmetered intelligence on your own machine. Open weights available now on @huggingface, run it locally on your favorite framework or try it for free on @opencode or @pidotdev.
Cool to see a bunch of project examples being done by our local model Flint.
Richard Socher retweeted
Introducing Flint, @Recursive_SI’s first model: a fast local coding model built with our automated research system, the Eureka Machine.
• Best-in-class fast local coding
• Up to 2x faster than comparable models: 125+ tokens/sec on a MacBook Pro, 150+ on RTX Spark
• Built from scratch in under 4 months
Private, unmetered intelligence on your own machine. Open weights available now on @huggingface, run it locally on your favorite framework or try it for free on @opencode or @pidotdev.
"I used Flint to build a bunch of programs and found it to be both fast and very capable. In my experience, models of this size can be annoyingly error-prone or slow compared to Claude or Codex, but Flint nailed my coding requests every time." — @willknight, @WIRED
wired.com/story/recursive-se…
Introducing Flint, @Recursive_SI’s first model: a fast local coding model built with our automated research system, the Eureka Machine.
• Best-in-class fast local coding
• Up to 2x faster than comparable models: 125+ tokens/sec on a MacBook Pro, 150+ on RTX Spark
• Built from scratch in under 4 months
Private, unmetered intelligence on your own machine. Open weights available now on @huggingface, run it locally on your favorite framework or try it for free on @opencode or @pidotdev.
Richard Socher retweeted
Recursive makes the case that the benefits of using AI to build AI outweigh the risks. buff.ly/xsPUDpl
Richard Socher retweeted
1/ From OpenAI's Navier–Stokes work to AlphaEvolve's faster algorithms, AI research is getting very real.
Now, with Recursive Flint, we can experience that magic locally on our own laptops 💻
The whole research loop runs locally with a selection of tasks! Code + demos 🧵
Flint is also a real-world test of the Eureka Machine, our system for automated AI research.
It runs the full research loop: ideation, implementation, and validation. Then it uses the results to decide what to try next. We used it in nearly every part of how Flint was built.
This is what allowed our very small team to build this model in <4 months.
Its contributions include a sparser model shape, low-bit training that shrank Flint by 71% with almost no loss in quality and custom kernels.
Try Flint today.
Read the blog: recursive.com/updates/introd…
Download on Hugging Face: huggingface.co/RecursiveSI/f…
Run Flint locally with @ollama, @lmstudio, @UnslothAI, @exolabs, @Nativ_AI, @sgl_project, @vllm_project and llama.cpp or try it for free on @opencode and @pidotdev
Richard Socher retweeted
Dr. @RichardSocher, CEO of Recursive, tells @JonKrohnLearns Krohn how AI could deliver a century of scientific breakthroughs in a decade, why recursive self-improvement may arrive within a year without an overnight takeoff, and which software and jobs will endure.
Watch the episode here: superdatascience.com/1037
Richard Socher retweeted
In my latest Ground Level AI podcast, @RichardSocher puts his probability of AI wiping out humanity at zero, arguing that extinction scenarios give AI “near magical abilities” to attack—but underestimate our ability to defend.
In the full episode (link in comments), we also discuss his conversations—and disagreements—with Geoffrey Hinton and Yoshua Bengio about AI extinction risk.
Richard Socher retweeted
"The p(doom) probability of existentially wiping out humanity is zero from AI.”
@Recursive_SI CEO @RichardSocher told #FortuneAIWeekly that AI doomsday scenarios “are so unrealistic that they're like borderline ridiculous.” bit.ly/4s07nJi
Richard Socher retweeted
@RichardSocher tells me & @beafreyanolan why his p(doom)=0. He also explains how close he thinks his new company @Recursive_SI is to automating scientific discovery. This was a great episode of @FortuneMagazine's Fortune AI Weekly. Check it out here: youtu.be/nutfdSHvLZI?si=WHb6…
When I saw some mathematicians complain about many conjecture being proven by AI, I couldn't help but think to myself...
I hope the medical biologists aren't going to be unhappy when AI cures most diseases.
Richard Socher retweeted
Humanity has problems that technology can solve well, @youdotcom and @Recursive_SI co-founder and CEO @RichardSocher tells @inafried.
“I believe that when we can automate research, first for AI but eventually also for all other kinds of sciences, humanity will massively benefit from us." #AxiosLive
Richard Socher retweeted
In reinforcement learning, inference is part of the training loop. Every checkpoint used to mean a redeploy, and the trainer waited.
CoreWeave RL Rollouts load the new weights into a live deployment without touching in flight requests, about 15x faster than a redeploy cycle.
@nvidia and @youdotcom used it to post train Nemotron 3.5 Lightning for web search and lifted BrowseComp accuracy from 36.97% to 45.45% while cutting tool calls by 30.24%.
Full breakdown here: crwv.co/utdYc
Richard Socher retweeted
Moonshots episode #299 with @RichardSocher of @recursive_si and @youdotcom is out. youtube.com/watch?v=Blyb1D92…
