GPT-6 Luna (High): benchmark results

Provider: OpenAI. Released 2026-09-22. Access: API.

Aggregate Rating 1760 ± 10, rank #224 of 2178 rated models, from 168 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Voice Standard v0.1.3 (claude-opus-5-5; 120s) - Information Extraction93Published information_extraction task_quality (1-100), canon100
Voice Standard v0.1.3 (gpt-5.6-sol; 900s) - Information Extraction89Published information_extraction task_quality (1-100), canon100
Voice Standard v0.1.3 (gpt-5.6-sol; 30s) - Information Extraction94Published information_extraction task_quality (1-100), canon99
o11y-bench - Pass@3100Tasks passed on at least one of three attempts, Pass@3 (%)96.4
Voice Standard v0.1.3 (gpt-5.6-sol; 120s) - Information Extraction94Published information_extraction task_quality (1-100), canon96.2
Voice Standard v0.1.3 (gpt-5.6-sol; 30s) - Summarization100Published summarization task_quality (1-100), canonical v0.196.1
Voice Standard v0.1.3 (claude-opus-5-5; 30s) - Text Correction96Published text_correction task_quality (1-100), canonical v095.3
Voice Standard v0.1.3 (claude-opus-5-5; 3600s) - Summarization90Published summarization task_quality (1-100), canonical v0.194.6
Voice Standard v0.1.3 (gpt-5.6-sol; 900s) - Summarization97Published summarization task_quality (1-100), canonical v0.192.2
Voice Standard v0.1.3 (claude-opus-5-5; 900s) - Summarization89Published summarization task_quality (1-100), canonical v0.190.2
Voice Standard v0.1.3 (gpt-5.6-sol; 120s) - Dictation97Published dictation task_quality (1-100), canonical v0.1.3, 89.4
Voice Standard v0.1.3 (claude-opus-5-5; 3600s) - Information Extraction81Published information_extraction task_quality (1-100), canon88.3

Interactive version: theaggregate.ai/model?slug=gpt-6-luna-high · How It Works · Data refreshed daily, snapshot 2026-11-07.