Week 45, 2026 standings1 Claude Opus 5.5 962 Gemini 4 Carbon 953 GPT-6 Astra 934 GPT-6.1 Sol 935 Claude Sonnet 5.5 926 Gemini 4 Argon 917 Claude Fable 5.1 908 Muse Wonder 899 Kimi K3.1 Pro 8810 Claude Haiku 5.5 87Week 45, 2026 standings1 Claude Opus 5.5 962 Gemini 4 Carbon 953 GPT-6 Astra 934 GPT-6.1 Sol 935 Claude Sonnet 5.5 926 Gemini 4 Argon 917 Claude Fable 5.1 908 Muse Wonder 899 Kimi K3.1 Pro 8810 Claude Haiku 5.5 87

Updated for Week 45, 2026

Claude Opus 5.5 vs GPT-6 Sol

Pick Claude Opus 5.5 for agentic coding. Pick GPT-6 Sol for daily work in ChatGPT.

	Claude Opus 5.5	GPT-6 Sol
Maker	Anthropic	OpenAI
Released	September 22, 2026	September 22, 2026
Price / 1M tokens	$4 in / $20 out	$2 in / $10 out
Open weights	No	No
Power Rankings rank	#1 (96)	—
Coding rank	#1 (98)	—
Best for	Agentic coding Long, multi-step knowledge work Writing and analysis	Daily work in ChatGPT Coding in Codex Balanced cost vs. quality
Strengths	Led six of the nine benchmarks Anthropic published at launch 66.4% on Terminal-Bench 4.0 at extra-high effort Much cheaper than the Fable/Mythos tier	Half the price of GPT-5.6 Sol's promo rates Fewer factual mistakes Available in ChatGPT, Codex, the API and GitHub
Watch out for	Trails GPT-6 Astra on AutomationBench and Terminal-Bench-Science Higher effort settings cost more tokens	Less reasoning headroom than Astra Brand-new; independent testing still coming in

The short version

Claude Opus 5.5: Anthropic's new default flagship. Opus 5.5 replaced Opus 5, costs 20% less, and at launch took the top spot on the independent Artificial Analysis Intelligence Index.

GPT-6 Sol: OpenAI's everyday workhorse. Sol brings Astra-style improvements to a much cheaper tier, and OpenAI says it makes about half as many mistakes as GPT-5.6 Sol.

Rankings change every week. Check the latest standings or swap in a different model.