Interesting paper I hadn't seen, the 'Densing Laws' of LLMs: they are getting twice as capable for the same size model every 3.3 months.

Densing Law of LLMs

Qwen 3 released today may be an emphatic continuation of the trend. Need to play with the models more to verify, but the benchmark numbers are... Staggering. Like 4 billion handily beating a 72 billion model from less than a year ago

Qwen

Qwen3: Think Deeper, Act Faster

QWEN CHAT GitHub Hugging Face ModelScope Kaggle DEMO DISCORD Introduction Today, we are excited to announce the release of Qwen3, the latest additi...

o3 isn't as good as I hoped, but it's still an increment in the SOTA. 69% on SWE-Bench Verified! The regression line over the past 2 years still points to 100‰ by year end! Frankly I think the real story is how cheaply Gemini 2.5 is delivering 64% on SWE-Bench Exciting times! Coding with Gemini 2.5 is so satisfying, a big step up from deepseek V3.1, which is what I was using before. #ai #llm #o3