Skip to content
All models

GPT-5.6 Luna

Legacy

OpenAIopenai/gpt-5.6-luna

Compare
Intelligence
Low50.5
Cost
Low$0.17
Blended, USD per 1M tokens
Value
High60.3
Counted at $0.14 per 1M tokens, from the measured run cost

How the intelligence score is built

How the intelligence score is built
MetricRawLeaderNormalisedWeight
AA Intelligence Indexgpt-5-6-luna23.864.636.821.7%
Arena textGPT-5.6 Luna (xHigh)1431.011538.2864.221.7%
Arena WebDevNot listed1835.76No data21.7%
Arena AgentNot listed0.1996No data21.7%
LiveBenchNot listed87.15No data13.3%
Intelligence
50.5

Intelligence is the weighted mean of the normalised leaderboard scores above. This model is on 2 of the 5 leaderboards. The ones it is not on are left out of the mean, not counted as 0.

History

History for this model starts on 2026-11-11. The chart appears once a second day is recorded.

Changes

No changes since 2026-11-11.

All changes