GPT-6 Luna

GPT-6 Luna is a reasoning model from OpenAI in the GPT-6 family. 12 benchmarks count toward its score, in 5 categories.

availableShows if the model has enough results for an index.
IndexOverall score out of 100.68.4 ±6.7
CoverageShare of the index weight with results.80%
SpeedOutput tokens per second.
Input / 1MUS dollars per 1M input tokens.$0.1
Output / 1MUS dollars per 1M output tokens.$0.5
ContextMaximum tokens in one request.1.05M
EloLMArena rating and rank.N/A

The index is a score out of 100. The ± range shows how much it can change.

CapabilitiesScore per category, out of 100.

Out of 100
AgenticMulti-step tasks with tools.
70.5
CodingCode writing and repair.
68.6
ReasoningLogic problems and puzzles.
72.5
MultimodalTasks with images and text.
59.4
KnowledgeFacts and expert knowledge.
68.7
MultilingualTasks in many languages.
N/A
InstructionTasks with strict rules in the prompt.
N/A
MathMath problems.
N/A

Results

12 counted
BenchmarkThe test name.CategoryThe capability that the test measures.ResultThe score from the publisher.IndexThis result as a score out of 100.RunThe settings of the run.DateDate of the result.Published byThe source of the result.
Artificial Analysis Long Context ReasoningReasoning83.3%65.9Artificial Analysis
Artificial Analysis MMMU-ProMultimodal75.5%59.4Artificial Analysis
DeepSWEAgentic66.6%72.2Datacurve AI
HealthBench Professional raw scoreKnowledge61.2%Anthropic
HealthBench ProfessionalKnowledge60.8%Rebecca Soskin Hicks et al.
Artificial Analysis SciCodeCoding54.6%68.6Artificial Analysis
HealthBench length-adjusted scoreKnowledge54.5%Anthropic
HealthBench raw scoreKnowledge50.0%Anthropic
Artificial Analysis Omniscience AccuracyKnowledge43.8%68.1Artificial Analysis
GDPval-AA normalizedAgentic43.4%68.9Artificial Analysis
Artificial Analysis Humanity's Last ExamKnowledge38.5%66.3Artificial Analysis
Artificial Analysis Intelligence IndexKnowledge37.3%69.0Artificial Analysis
HealthBench HardKnowledge31.4%71.5Meta AI
Artificial Analysis GDP.pdfAgentic20.4%71.5Artificial Analysis
Critical Physics TasksReasoning19.4%79.1Artificial Analysis
ExploitGymAgentic11.6%69.6Zhun Wang et al.

12 benchmarks count, from 12 of 16 results. A grey row does not count. Too few models took that benchmark.

Sources

BenchLM benchmark aggregationUsed with attribution; per-benchmark results credited to their original authors

Same level, lower price

Qwen3.8-Flash-Next66.4 · Freedots3-note Preview65.5 · Free

More from OpenAI

GPT-6 Astra83.5GPT-5.6 Sol76.7GPT-5.6 Terra73.4GPT-6 Sol76.9GPT-5.572.3