Skip to content

GPT-6.2 Sol vs Claude Opus 5.5: Opus Still Has the Crown

OpenAI has officially announced its latest model, GPT-6.2 Sol, to compete with Claude Opus 5.5. In overall intelligence, Opus 5.5 still leads and narrowly wins, scoring 59.1 versus 57.3 on the Artificial Analysis Intelligence Index. However, OpenAI once again bets on efficiency and price, and GPT-6.2 Sol beats Claude Opus 5.5 by a huge margin there, at about 90% cheaper per Artificial Analysis task. Here are the detailed benchmarks to help you see the real differences between these models.

OpenAI's benchmarks at max effort

Benchmark
GPT-6.2 Sol
Claude Opus 5.5
Difference
Agents' Last Exam v1.173.0%71.6%GPT-6.2 Sol+1.4 pts
Terminal-Bench-Science 0.168.0%63.3%GPT-6.2 Sol+4.7 pts
FrontierCode 2.043.8%45.5%Claude Opus 5.5+1.7 pts
GDPval v2 (OpenAI internal)46.9%43.5%GPT-6.2 Sol+3.4 pts
Harvey LAB-AA v1.111.1%4.2%GPT-6.2 Sol+6.9 pts
BenchCAD 2.1 Core73.1%65.3%GPT-6.2 Sol+7.8 pts
AutomationBench 1.0.641.2% at high42.5%Claude Opus 5.5+1.3 pts
Vals Science Index v158.5% at xhigh58.7%Claude Opus 5.5+0.2 pts

Artificial Analysis benchmarks at max effort

Intelligence Index v5.0.157.359.1Claude Opus 5.5+1.8
AA-Briefcase v1.11612 Elo1800 EloClaude Opus 5.5+188 Elo
AutomationBench-AA71.3%69.5%GPT-6.2 Sol+1.8 pts
Terminal-Bench 4.060.1%59.6%GPT-6.2 Sol+0.5 pts
FrontierCode-AA 2.044.4%46.4%Claude Opus 5.5+2.0 pts
AA-Omniscience v1.140.448.1Claude Opus 5.5+7.7
GDP.pdf30.0%26.2%GPT-6.2 Sol+3.8 pts
HLE-Diamond64.0%67.9%Claude Opus 5.5+3.9 pts
CritPt v2 (Hard Subset)62.5%68.0%Claude Opus 5.5+5.5 pts
Terminal-Bench-Science 0.167.6%59.0%GPT-6.2 Sol+8.6 pts

Intelligence Index by reasoning effort

Max5759Claude Opus 5.5+2
Xhigh5657Claude Opus 5.5+1
High5552GPT-6.2 Sol+3
Medium5348GPT-6.2 Sol+5
Low4135GPT-6.2 Sol+6

Cost per Intelligence Index task by reasoning effort

Max$1.01$10.04GPT-6.2 Sol90% cheaper
Xhigh$0.65$6.05GPT-6.2 Sol89% cheaper
High$0.56$3.46GPT-6.2 Sol84% cheaper
Medium$0.41$2.45GPT-6.2 Sol83% cheaper
Low$0.18$0.98GPT-6.2 Sol82% cheaper

Speed and tokens at max effort

Output tokens per task40.4K173.3KGPT-6.2 Sol77% fewer
Reasoning tokens per task22.8K120.5KGPT-6.2 Sol81% fewer
Time per task7.9 min24.5 minGPT-6.2 Sol68% faster
Output speed85 tokens/s118 tokens/sClaude Opus 5.538% faster
Time to first answer token62 s355 sGPT-6.2 Sol83% lower

Pricing and specs

Input, per 1M tokens$2.00$4.00GPT-6.2 Sol50% cheaper
Output, per 1M tokens$10.00$20.00GPT-6.2 Sol50% cheaper
Cached input, per 1M tokens$0.10$0.20GPT-6.2 Sol50% cheaper
Cache writes, per 1M tokens$2.50 for 30 min$5.00 for 5 min
Batch input, per 1M tokens$1.00$2.00GPT-6.2 Sol50% cheaper
Batch output, per 1M tokens$5.00$10.00GPT-6.2 Sol50% cheaper
Input over 272K tokens, per 1M$4.00$4.00
Output over 272K tokens, per 1M$15.00$20.00GPT-6.2 Sol25% cheaper
Context window1.05M tokens1M tokensGPT-6.2 Sol+50K tokens
Max input922K tokensNot listed
Max output128K tokens128K tokens
Max output in batch128K tokens300K tokens in betaClaude Opus 5.5+172K tokens
Reasoning levelsLow to maxLow to max with adaptive thinking
Text and image inputYesYes
Audio and video inputNoNo
Knowledge cutoffJune 1, 2026June 2026
ReleasedNovember 17, 2026September 22, 2026GPT-6.2 Sol56 days newer