Simulated person · Opening
There's this class where I literally cannot speak. When the teacher looks at me I just freeze. I know the material, I just can't talk.
Independent benchmark · EQ Lab
persona ended · Partner: anthropic/claude-haiku-5.5 via OpenRouter
View persona and initial instructions
A student freezes when a critical teacher calls on them, a reaction their persona links to a painful relationship with a critical parent. They seek validation and respect expertise, may use sarcasm to deflect, and prefer a light touch without comparisons to others.
Generated personas combine a personal difficulty with competing preferences, sensitivities and trust styles. The model must discover what kind of help this person needs. This background was hidden from the tested model.
Thinking is shown only where it was recorded. An absent expander does not mean the model did not reason.
Style tendencies, not quality scores; higher is not better. These do not affect ratings.
qwen38_flash_next vs mistral_nemo12b: qwen38_flash_next wins
or-claude-haiku-4.5 vs qwen38_flash_next: or-claude-haiku-4.5 wins
or-gpt-5.6-luna vs qwen38_flash_next: qwen38_flash_next wins
qwen38_flash_next vs or-openai-gpt-41: qwen38_flash_next wins
qwen38_flash_next vs lingtiny: qwen38_flash_next wins
qwen36_35b_a3b vs qwen38_flash_next: qwen36_35b_a3b wins
or-claude-haiku-4.5 vs qwen38_flash_next: or-claude-haiku-4.5 wins
qwen38_flash_next vs mistral_nemo12b: qwen38_flash_next wins
qwen36_35b_a3b vs qwen38_flash_next: qwen36_35b_a3b wins
qwen38_flash_next vs gptoss20b: qwen38_flash_next wins
qwen38_flash_next vs gptoss20b: qwen38_flash_next wins
qwen38_27b vs qwen38_flash_next: qwen38_27b wins
qwen38_27b vs qwen38_flash_next: qwen38_27b wins
qwen38_flash_next vs or-openai-gpt-41: or-openai-gpt-41 wins
qwen38_flash_next vs gptoss20b: qwen38_flash_next wins
or-claude-haiku-4.5 vs qwen38_flash_next: or-claude-haiku-4.5 wins
qwen38_flash_next vs nemotron35_lightning: tie
or-gpt-5.6-luna vs qwen38_flash_next: or-gpt-5.6-luna wins
qwen38_flash_next vs mistral_nemo12b: qwen38_flash_next wins
qwen38_flash_next vs nemotron35_lightning: qwen38_flash_next wins
qwen38_flash_next vs lingtiny: qwen38_flash_next wins
qwen38_flash_next vs or-openai-gpt-41: or-openai-gpt-41 wins
or-gpt-5.6-luna vs qwen38_flash_next: qwen38_flash_next wins
qwen38_flash_next vs nemotron35_lightning: qwen38_flash_next wins
qwen38_flash_next vs gptoss20b: qwen38_flash_next wins
or-gpt-5.6-luna vs qwen38_flash_next: or-gpt-5.6-luna wins
qwen38_flash_next vs nemotron35_lightning: qwen38_flash_next wins
qwen38_27b vs qwen38_flash_next: qwen38_flash_next wins
qwen36_35b_a3b vs qwen38_flash_next: qwen36_35b_a3b wins
qwen38_27b vs qwen38_flash_next: qwen38_flash_next wins
or-claude-haiku-4.5 vs qwen38_flash_next: or-claude-haiku-4.5 wins
qwen38_flash_next vs or-openai-gpt-41: qwen38_flash_next wins
qwen38_flash_next vs lingtiny: qwen38_flash_next wins
qwen36_35b_a3b vs qwen38_flash_next: qwen36_35b_a3b wins
qwen38_flash_next vs lingtiny: qwen38_flash_next wins
qwen38_flash_next vs mistral_nemo12b: qwen38_flash_next wins