hybrid-opus-haiku-for-researchclaw-stay-anthropic

AutoResearchClaw’s 23 pipeline stages should use Opus 4.6 for 8 critical reasoning stages (hypothesis generation, paper writing, peer review, experiment design) and Haiku 4.5 for 15 routine stages. Do NOT use GPT-4o-mini or external LLMs — stay within the Anthropic ecosystem to avoid external API dependencies and maintain consistency. AutoResearchClaw natively supports primary + fallback model configuration, making this pattern straightforward to wire.