Reddit user: Opus 5.5 silently falls back to Opus 5 on nearly every prompt
fishcat_catfish · reddit · 2026-09-29
A Reddit user reports that Opus 5.5 almost never actually serves their prompts — requests silently fall back to Opus 5 without any explanation, even for ordinary scientific work. They note Anthropic's Astra burns through limits faster and benchmarks worse, but at least runs. The post highlights recurring complaints about models being advertised but not actually available.
More from Models
- Is Anthropic 'really back'? User urges running Opus 5.5 + Sonnet 5.5 swarms — IndraVahan · 2026-09-29
- A week with Sonnet 5.5: ask it for ideas and it builds the thing instead — every · 2026-09-29
- Aether AI releases 16B CausalWM, a world model that reasons about causality before generating future frames — jiqizhixin · 2026-09-29
- Early Sonnet 5.5 impressions: ~5x faster bug fixes at high effort — burkov · 2026-09-29
- Why 4o feels different: thread argues native omni training, not capability, shapes model personality — RileyRalmuto · 2026-09-29
- Sonnet 5.5 effort settings make no difference in 15-task coding test: 9/15 at low, medium and high — every · 2026-09-29