andyk

Pplx and Dbrx founder giving $1M to first OSS AI that gets 90% on SWE-bench

#1 agent on swe-bench wrote 7% of its own code

Headless Terminal – making terminals easy for LLMs to use

AI benchmarks should be like unit tests

OpenAI Dropped Work on New 'Arrakis' AI Model in Rare Setback

SlimPajama: A 627B token cleaned and deduplicated version of RedPajama

The HTTP QUERY Method specification