Benchmark charts on model card: https://huggingface.co/01-ai/Yi-1.5-34B-Chat#benchmarks
Yi 34b with results similar to Llama 3 70b and Mixtral 8x22b
Yi 6b and 9b with results similar to Llama 3 8b
Yi 34b with results similar to Llama 3 70b and Mixtral 8x22b
Yi 6b and 9b with results similar to Llama 3 8b
The "will it tell me how to make meth" stuff is a huge source of noise, which you could argue is digging for refusals which can be annoying, and the benchmark claims to filter out... but in reality a bunch of the refusals are soft refusals that don't get caught, and people end up downvoting the model that's deemed "corporate".
Honestly the fact that any closed source model with guardrails can even place is a miracle, in a proper benchmark the honest to goodness gap between most closed source models and open source models would be so large it'd break most graphs.
There aren't nearly as many 3.5 level models as the leaderboard implies for example.
LLM benchmarks are horribly broken. IMHO there is better signal in just looking at parameter counts.