https://rdolmedo.github.io/Ricardo Olmedo: final-year PhD student at the Max Planck Institute for Intelligent Systems, working on LLM benchmarking & specialization.https://rdolmedo.github.io/blogs/minicoder.htmlhttps://rdolmedo.github.io/blogs/minicoder.htmlTue, 30 Sep 2025 00:00:00 +0000Two distilled models (1.7B & 4B) + 400k training trajectories from Qwen 3 Coder. Download the models & dataset here.https://rdolmedo.github.io/blogs/arena.htmlhttps://rdolmedo.github.io/blogs/arena.htmlFri, 02 May 2025 00:00:00 +0000Releasing all Arena data would make LMArena a fairer leaderboard. It would also force us to reflect on whether we actually care who is at the top.