Llama 4 underperforms: a benchmark against coding-centric modelsrootly.com5 points·sylvainkalache··0 commentsOpen articleSaveView on HN