Compare Models

Claude Opus 5.5 (Adaptive Reasoning, Max Effort, Default Fallback) vs GPT-6 Astra (max)

A metric-by-metric comparison using the latest available benchmark, runtime, and pricing values. Claude Opus 5.5 (Adaptive Reasoning, Max Effort, Default Fallback) wins more comparable metrics.

Open Full Benchmarks Comparison Builder

Claude Opus 5.5 (Adaptive Reasoning, Max Effort, Default Fallback): 8
GPT-6 Astra (max): 0
Ties: 0
Metrics compared: 12

Compare Models

Metric	Claude Opus 5.5 (Adaptive Reasoning, Max Effort, Default Fallback)	GPT-6 Astra (max)
Artificial Analysis Intelligence Index	57.6	52.7
Artificial Analysis Coding Index	n/a	76.9
LiveBench Mathematics	97.1%	96.8%
GPQA	n/a	96.1%
Humanity's Last Exam	61.4%	54.7%
SciCode	66.9%	56.5%
Output Speed	n/a	54.3 tok/s
Time to First Token	n/a	203.02 s
Blended Price	$8.00 / 1M	$20.00 / 1M
Input Price	$4.00 / 1M	$10.00 / 1M

Data & Evidence

Claude Opus 5.5 (Adaptive Reasoning, Max Effort, Default Fallback)

Anthropic

Open model profile

GPT-6 Astra (max)

OpenAI

Open model profile