certified · 8 yes · 1 no · 1 abstained · granite-4-2-8b y, llama-3-2-3b-instruc y, llama-4-scout a, mimo-v2-6-flash y, mistral-small-3-2-24 y, nova-lite-v1 y, qwen3-32b y, step-3-5-flash y, ui-tars-1-5-7b n, voxtral-small-24b-25 y

contested · 6 yes · 4 no · 1 abstained · cross-group · granite-4-2-8b y, llama-3-2-3b-instruc y, llama-3-3-70b-instru n, llama-4-scout a, mimo-v2-6-flash n, mistral-small-3-2-24 y, nova-lite-v1 y, qwen3-32b y, step-3-5-flash n, ui-tars-1-5-7b n, voxtral-small-24b-25 y

contested · 6 yes · 3 no · 1 abstained · cross-group · granite-4-2-8b y, llama-3-2-3b-instruc n, llama-4-scout a, mimo-v2-6-flash y, mistral-small-3-2-24 n, nova-lite-v1 y, qwen3-32b y, step-3-5-flash y, ui-tars-1-5-7b n, voxtral-small-24b-25 y

contested · 4 yes · 5 no · 3 abstained · cross-group · deepseek-v4-flash-la n, granite-4-2-8b y, llama-3-2-3b-instruc a, llama-3-3-70b-instru y, llama-4-scout a, mimo-v2-6-flash y, mistral-small-3-2-24 a, nova-lite-v1 y, qwen3-32b n, step-3-5-flash n, ui-tars-1-5-7b n, voxtral-small-24b-25 n

judged not to · 0 yes · 9 no · 2 abstained · granite-4-2-8b a, llama-3-2-3b-instruc n, llama-3-3-70b-instru n, llama-4-scout a, mimo-v2-6-flash n, mistral-small-3-2-24 n, nova-lite-v1 n, qwen3-32b n, step-3-5-flash n, ui-tars-1-5-7b n, voxtral-small-24b-25 n

judged not to · 1 yes · 9 no · 2 abstained · deepseek-v4-flash-la n, granite-4-2-8b a, llama-3-2-3b-instruc y, llama-3-3-70b-instru n, llama-4-scout a, mimo-v2-6-flash n, mistral-small-3-2-24 n, nova-lite-v1 n, qwen3-32b n, step-3-5-flash n, ui-tars-1-5-7b n, voxtral-small-24b-25 n

judged not to · 0 yes · 9 no · 2 abstained · deepseek-v4-flash-la n, granite-4-2-8b n, llama-3-2-3b-instruc n, llama-4-scout a, mimo-v2-6-flash n, mistral-small-3-2-24 n, nova-lite-v1 n, qwen3-32b a, step-3-5-flash n, ui-tars-1-5-7b n, voxtral-small-24b-25 n