Why AlphaEvolve matters This isn’t just about math. AlphaEvolve shows: – LLMs + evaluators can discover new knowledge
– Code generation can evolve into true algorithm innovation
– AI agents can now tackle hard science problems autonomously
LLMS
-
AlphaEvolve: LLMs and Evaluators Discover New Algorithms
By
–
-
AlphaEvolve: Darwin-Inspired Code Evolution with Gemini Models
By
–
Built with Gemini: AlphaEvolve uses a model ensemble: – Gemini Flash for rapid idea generation
– Gemini Pro for deeper, high-quality code suggestions Each candidate program is evaluated, scored, and evolved Darwin meets compilers. -
AlphaEvolve: LLM-Powered Evolutionary Coding Agent for Algorithm Optimization
By
–
AlphaEvolve: It’s an evolutionary coding agent powered by large language models (LLMs) + automated evaluators. It doesn’t just write code it evolves algorithms over time, scoring and mutating them until it finds breakthroughs.
-
o3 Model Image Critique Test After Three Years
By
–
Tried this on a lark 3 years later: Textual discussion is just o3 given the original image in same thread and told “Critique it.”
-
Multiplayer LLM Access for Organizational Learning
By
–
I very much understand that public by default would not work for much LLM use. But I think you could greatly increase rate of learning within e.g. an org by making a primary way LLMs are accessed a surface which explicitly considered multiplayer use, maybe by default.
-
Sonnet pricing analysis: caching benefits versus marginal costs
By
–
I certainly wouldn't say it's clearly worth the extra cost, at least. (Especially since I use caching a lot, which is a 90% discount for Sonnet.)
-
Why Language Models Build Rich Knowledge Hierarchies
By
–
…which is literally why I created LLMs. I explained in the original videos that the entire reason was because language models, to be good at their job, must build a rich hierarchy of abstractions and capabilities and implicit world-knowledge. Turns out it works! 😀
-
Exploring Alternative Visualizations for Transformer Internal Mechanisms
By
–
surely there are other ways to visualize the internal mechanisms of transformers
-
R1 Open Source Model Rivals o3 Mini Performance
By
–
Of course Sonnet is superior. To me, this new R1 is closer to the o3 Mini, which is pretty wild considering it’s open source.
-
Claude Sonnet 4 outperforms for coding and writing tasks
By
–
Yeah sonnet 4 definitely better for me. Using it mainly for coding, but also for various writing and admin tasks. It does a better job of understanding and sticking to what I’m asking.