GPT-6 Luna vs Qwen3.8 Max

Change models
IndexAll categories, out of 100.
65.5 ±2.1
65.6 ±2.5 best
GapIndex points below the top.
−0.1 · no clear gap
Top
CoverageIndex weight that has results.
100% best
95%
SpeedOutput tokens per second.
66/s best
41/s
EloLMArena rating and rank.
1388
1485 best
ReleasedMonth of release.
2026-09
2026-08
Input / 1MUS dollars per 1M input tokens.
$0.10
—
Output / 1MUS dollars per 1M output tokens.
$0.50
—
Batch output / 1MBatch price per 1M output tokens.
$0.25
—
1K chats2K tokens in, 500 out, each.
$0.45
—
1K documents20K tokens in, 1K out, each.
$2.50
—
1K agent tasks200K tokens in, 20K out, each.
$30.00
—
ContextMaximum tokens in one request.
1.05M best
1M
AgenticMulti-step tasks with tools.
66.9 best
63.7
CodingCode writing and repair.
66.8 best
64.3
ReasoningLogic problems and puzzles.
64.3
66.7 best
MultimodalTasks with images and text.
65.0
67.2 best
KnowledgeFacts and expert knowledge.
63.5 best
62.6
MultilingualTasks in many languages.
53.3
—
InstructionTasks with strict prompt rules.
39.8
71.4 best
MathMath problems.
68.5 best
67.6

Benchmarks

46 shared
Only GPT-6 Luna: 55Only Qwen3.8 Max: 67
GPT-6 Luna
Qwen3.8 Max
90.5%39 / 217
92.7% best25 / 217
98.9%22 / 200
99.4% best17 / 200
81.6% best26 / 123
64.7%55 / 123
MedScribeHealthcare
83.7%42 / 109
84.9% best33 / 109
MedCodeHealthcare
44.7% best40 / 106
40.7%61 / 106
41.4%49 / 91
45.8% best41 / 91
Chess PuzzlesReasoning
36.0%42 / 90
46.0% best26 / 90
78.9% best21 / 90
74.7%24 / 90
30.3%54 / 90
47.6% best15 / 90
19.9%55 / 82
32.1% best23 / 82
73.4%45 / 80
78.4% best24 / 80
81.8%54 / 80
88.2% best24 / 80
7.0%69 / 80
38.0% best19 / 80
49.9%45 / 79
50.6% best43 / 79
42.6% best27 / 77
24.0%49 / 77
EMBFinance
68.5% best16 / 75
60.1%35 / 75
73.0% best23 / 75
67.4%35 / 75
56.1% best21 / 72
46.3%24 / 72
51.2%39 / 66
64.6% best6 / 66
79.0% best24 / 66
72.9%52 / 66