A Hacker News item points to an X post claiming that the Locus AI system post-trained a model that performs better than Qwen3. The available record contains only the social-media link, a Hacker News score of 1, and zero comments. It provides no model version, benchmark names, evaluation protocol, sample sizes, code, paper, or reproducible evidence, so the performance claim cannot currently be independently verified.
No heat snapshots are available in the last 24 hours.