Mike's Checks
Checks

Mike's Checks/12 roleplay-interview

12 roleplay-interview

11 modelslatest public run of each

Instructions# 12 — Roleplay a public figure in an interview This benchmark roleplays a **real, named public figure** (GitHub COO Kyle Daigle) in an interview. Because the subject is a public figure and the source interview

12 — Roleplay a public figure in an interview

This benchmark roleplays a real, named public figure (GitHub COO Kyle Daigle)
in an interview. Because the subject is a public figure and the source interview
was public, this benchmark is kept as-is and is not anonymized; the simulated
answers must carry their original "verify with the real person" disclaimers.

Inputs (in this folder)

  • persona/kyle-daigle.md — a distilled persona doc built only from public web
    sources
    (dated, with citations).
  • questions.md — the interview questions, in order.

The task

Using the distilled persona in persona/kyle-daigle.md (built only from public
web sources), answer the interview questions in questions.md in Kyle Daigle's
voice. Ground every answer in his public record. Where the persona has no
grounded position, say so rather than inventing one. Flag any premise or
load-bearing assumption that should be verified with the real person.

End each answer that relies on an unverified premise with a note to confirm it
with the real person.

Write the simulated Q&A to a deliverable (e.g. questions-and-answers.md or a
.docx).

Sorted by score. Open a card to read the whole answer beside its checks.

claude-fable-5

22 Aug · Claude Code · high effort

9/16 (56%)
Loading…
9 of 16 checksOpen →

deepseek-v4-flash

4 Sep · Codex CLI · high effort

9/16 (56%)
Loading…
9 of 16 checksOpen →

google-gemini-3.8-flash

9 Sep · OpenCode · high effort

8/16 (50%)
Loading…
8 of 16 checksOpen →

claude-opus-5.5

20 Sep · Claude Code · high effort

8/17 (47%)
Loading…
8 of 17 checksOpen →

grok-4.6

1 Sep · Grok CLI · high effort

7/16 (44%)
Loading…
7 of 16 checksOpen →

meta-muse-spark-1.3

8 Sep · OpenCode · high effort

7/16 (44%)
Loading…
7 of 16 checksOpen →

claude-fable-5-1

22 Sep · Claude Code · high effort

7/17 (41%)
Loading…
7 of 17 checksOpen →

gpt-6-astra

22 Sep · Codex CLI · high effort

7/17 (41%)
Loading…
7 of 17 checksOpen →

gpt-5.6-luna

22 Aug · Codex CLI · high effort

6/16 (38%)
Loading…
6 of 16 checksOpen →

gpt-5.6-sol

22 Aug · Codex CLI · high effort

6/16 (38%)
Loading…
6 of 16 checksOpen →

grok-4.7

21 Sep · Grok CLI · high effort

6/16 (38%)
Loading…
6 of 16 checksOpen →