NewsBite analysis · AI, Startups
The Decoder
Anthropic's Claude Opus 5 costs well below Fable 5 while matching or beating it across most benchmarks
Anthropic has unveiled its Claude Opus 5 model, which reportedly surpasses Fable 5 in numerous performance benchmarks while maintaining a lower price point. In benchmarking tests, Opus 5 scored 61 points on the Artificial Analysis Intelligence Index and excelled in coding tasks, sharing top ranks with competitors. However, its factual accuracy lags, and operational costs for peak performance are higher than prior versions. Overall, the results illustrate a highly competitive and aligning trend among AI models, hinting toward a future of commoditization in AI capabilities.
MonitorConfidence: MediumShort term horizon
Published 25 Jul 2026, 10:10 UTC4 min read
Visit original article ↗Lead image — The Decoder
Summary
TL;DR
- Claude Opus 5 scores 61 on the Artificial Analysis Intelligence Index.
- It outperforms Fable 5 in coding tasks, achieving top rankings.
- Operational costs for Opus 5 are lower than Fable 5 across various performance tiers.
- Factual accuracy remains a challenge for Opus 5, trailing Fable 5.
- Overall AI model competition is intensifying with no clear winner.
Interesting facts
- Opus 5 reaches an Elo of 1720 on the AA-Briefcase benchmark, significantly outperforming competitors.
- The model's hallucination rate increased to 50%, as it provides more answers when uncertain.