From https://eqbench.com/creative_writing_longform.html it’s good at producing long paragraphs at a high quality, other models are quite hesitant to produce long paragraphs, and even on this benchmark site the judge model punishes long paragraphs.
Claude opus 4.7 is too expensive at the moment and recently you’ve only added closed source frontier models, which all have the same issues and make the tone variety feel bland.
Please authenticate to join the conversation.
Proposed
💡 Feature Request
1 day ago

Kiss butterfly
Get notified by email when there are changes.
Proposed
💡 Feature Request
1 day ago

Kiss butterfly
Get notified by email when there are changes.