Models favor their own APIs in code: Opus calls Claude API in 97% of implementations
santiviquez · x · 2026-11-17
In @santiviquez's API recommendation experiment, models showed clear self-preference when asked to write implementation code instead of giving verbal recommendations.
- Opus 5.5 called the Claude API in 58 of 60 implementations (97%), despite recommending Claude in only 10 of 150 verbal answers to the same prompts
- GPT-6 Astra called the OpenAI API in 27 of 60 implementations (45%), despite never recommending it verbally
- Search behavior also diverged: Opus searched in only 10 of 60 answers, Fable in 24, while both GPT models searched every time
More from Models
- 'Humans Also Read Linearly and Causally, So It's Not a Transformer Issue' — yacineMTB · 2026-11-17
- Fudan & OpenMOSS map the trade-off: long-context hybrids learn more, extrapolate less reliably — jiqizhixin · 2026-11-17
- Search behavior gap: Opus searched in 10/60 answers, GPT models searched every time — santiviquez · 2026-11-17
- 79% of 2,880 LLM API recommendations went to just four big-cloud services — santiviquez · 2026-11-17
- Hy4 Flash hands-on: DeepSeek-V4.1-level quality at ~1/15 the cost, 60% faster — IlovedamemesUwU · 2026-11-17
- Memory eval leaderboards are misleading: OmniMemEval mixes conditions and tunes on its own benchmark — Snoo-bedooo · 2026-11-17