Best LLMs for onboarding_chapter_prompt_generation autogenerated — DTP Benchmark
Models
Frontier on this task: GPT-5.5 at 9.18 / 10. Quality bar at 95%: 8.72.
point-estimate floor (CI low) · upper CI (less certain) · Bars sorted by blended cost; cheapest qualifier first. Greyed rows are MEDIUM+ models whose point estimate clears the bar but whose CI low does not.
Cost breakdown
| Model | Quality | Sample | Blended cost / call | Savings vs best | Mode |
|---|---|---|---|---|---|
| Kimi K2.6 Moonshot AI | 8.75 / 10 CI [8.59, 8.91] | n=86 · ranked | $0.070194 | 84% cheaper | batch |
| Claude Opus 4.7 Anthropic | 8.98 / 10 CI [8.76, 9.19] | n=75 · high | $0.363502 | 16% cheaper | batch |
| GPT-5.5 best OpenAI | 9.18 / 10 CI [9.06, 9.31] | n=78 · ranked | $0.433970 | (anchor) | batch |
Typical call shape for this task: 4466 input tokens → 28187 output tokens, EMA-tracked from production traffic. Blended cost = (in × in_price + out × out_price), rounded to 6 decimals.