Any good alternatives to Artificial Analysis?

Excerpt

So I used to use Artificial Analysis to compare models. The recent 61 score for Astra had me look into the results more granularly and I was very disappointed with what I found. Different pages reporting different scores for the same model on the same benchmark. Scores on a benchmark completely in contradiction with the benchmark providers verified results etc. In short, that site cannot be trusted as a source of data. Are there any good alternatives? A lot of the direct benchmarks (DeepSWE, ter