Claude Opus 5.5 matches GPT-6 Astra on intelligence at 4.5x lower cost per task
MaziyarPanahi · x · 2026-10-28
Latest Artificial Analysis data at max reasoning shows Claude Opus 5.5 (max/xhigh with fallback) topping the Intelligence Index at 52, with GPT-6 Astra (max/xhigh) right behind at 53 — effectively tied on intelligence (Healthcare & Medical Index: 52 vs 53) but Claude costs $0.72 per Intelligence Index task vs $3.26, a 4.5x gap. Other leaders: Celeris-1 is fastest at 1562 tokens/s, K2 Horizon 375B A23B follows at 929 t/s and is the cheapest at $0/M tokens; North Mini Code has the lowest latency (0.49s); Llama 4 Scout offers the largest 10M context window. The Intelligence Index v5.0 combines 9 evals including Terminal-Bench 4.0, FrontierCode 2.0, GDP.pdf and HLE-Diamond across 28 of 715 tracked models.
More from Models
- LiquidAI's LFM3-Omni-2B any-to-any model trends on Hugging Face — LiquidAI · 2026-10-28
- Founder: current AI models' writing is 'personally embarrassing' and fixes are comically underbaked — AndrewSchmidtFC · 2026-10-28
- Extrapolating Elo curves: end-of-2026 'GPT-6' would beat GPT-5.2 by ~370 Elo points — scaling01 · 2026-10-28
- Users report ChatGPT quietly brings back '5.6 Sol', which beats GPT-6 at audio transcription — tunerhd · 2026-10-28
- Google unveils Genie 4 world model, GPT-6.1 Terra lands at 50% lower API prices — testingcatalog · 2026-10-28
- Local Qwen3.8 prefill repro: 40.8% of samples complete to aliyuncs.com exfil URL — Nils_Reimers · 2026-10-28