Benchgen

ARC-C — Results

RankModelScore
1mimo-v2-5-pro0.972
2llama-3-1-405b-instruct0.969
3claude-3-opus0.964
4llama-3-1-70b-instruct0.948
5nova-2-pro0.948
6claude-3-sonnet0.932
7jamba-1-5-large0.93
8nova-2-lite0.924
9mistral-small-3-24b-base0.913
10phi-3-5-moe-instruct0.91
11claude-3-haiku0.892
12jamba-1-5-mini0.857
13phi-3-5-mini-instruct0.846
14phi-4-mini0.837
15llama-3-1-8b-instruct0.834
16qwen2-5-coder-32b-instruct0.705
17qwen2-5-32b-instruct0.704
18qwen2-72b-instruct0.689
19qwen2-5-14b-instruct0.673
20hermes-3-70b0.655
21gemma-3n-e4b-litert-preview0.616
22gemma-3n-e4b0.616
23qwen2-5-coder-7b-instruct0.609
24gemma-3n-e2b-litert-preview0.517
25gemma-3n-e2b0.517
A

ARC-C — Evaluations

1 phaseActive

No evaluations yet for this benchmark.