GPT-5.6 Luna vs GPT-6 Luna

Change models
IndexAll categories, out of 100.
65.5 ±2.1
65.5 ±2.1
GapIndex points below the top.
Top
Top
CoverageIndex weight that has results.
100%
100%
SpeedOutput tokens per second.
51/s
66/s best
EloLMArena rating and rank.
1431 best
1388
ReleasedMonth of release.
2026-07
2026-09
Input / 1MUS dollars per 1M input tokens.
$0.20
$0.10 best
Output / 1MUS dollars per 1M output tokens.
$1.20
$0.50 best
Batch output / 1MBatch price per 1M output tokens.
$0.60
$0.25 best
1K chats2K tokens in, 500 out, each.
$1.00
$0.45 best
1K documents20K tokens in, 1K out, each.
$5.20
$2.50 best
1K agent tasks200K tokens in, 20K out, each.
$64.00
$30.00 best
ContextMaximum tokens in one request.
1.05M
1.05M
AgenticMulti-step tasks with tools.
64.7
66.9 best
CodingCode writing and repair.
67.4 best
66.8
ReasoningLogic problems and puzzles.
66.3 best
64.3
MultimodalTasks with images and text.
65.8 best
65.0
KnowledgeFacts and expert knowledge.
62.1
63.5 best
MultilingualTasks in many languages.
53.1
53.3 best
InstructionTasks with strict prompt rules.
48.2 best
39.8
MathMath problems.
68.4
68.5 best

Benchmarks

70 shared
Only GPT-5.6 Luna: 32Only GPT-6 Luna: 31
GPT-5.6 Luna
GPT-6 Luna
39.5% best49 / 222
38.5%54 / 222
83.7% best10 / 217
83.3%13 / 217
91.6% best31 / 217
90.5%39 / 217
20.6% best29 / 214
19.4%31 / 214
EuroEval PolishMultilingual
54.2%50 / 209
56.0% best36 / 209
98.3%28 / 200
98.9% best22 / 200
EuroEval SwedishMultilingual
66.9% best15 / 195
61.2%31 / 195
EuroEval GermanMultilingual
52.6% best27 / 191
50.8%36 / 191
60.7% best27 / 182
60.6%29 / 182
EuroEval FrenchMultilingual
68.9% best10 / 178
62.7%29 / 178
EuroEval SpanishMultilingual
57.9% best19 / 176
52.1%38 / 176
EuroEval DutchMultilingual
59.3% best28 / 175
59.0%29 / 175
EuroEval ItalianMultilingual
60.0%30 / 174
62.1% best24 / 174
42.7% best42 / 164
39.6%54 / 164
48.0% best30 / 133
46.6%36 / 133
59.5% best43 / 131
59.3%44 / 131
88.0% best42 / 126
86.7%46 / 126
77.1%39 / 123
81.6% best26 / 123
60.2%38 / 122
61.8% best36 / 122
90.8%43 / 116
92.5% best40 / 116