GPT-6.2 Sol vs Muse Wonder

Change models
IndexAll categories, out of 100.
76.0 ±4.6 best
71.5 ±3.9
GapIndex points below the top.
Top
−4.5 · no clear gap
CoverageIndex weight that has results.
85%
85%
SpeedOutput tokens per second.
56/s
126/s best
EloLMArena rating and rank.
1414
1503 best
ReleasedMonth of release.
2026-11
2026-10
Input / 1MUS dollars per 1M input tokens.
$2.00 best
$4.00
Output / 1MUS dollars per 1M output tokens.
$10.00 best
$20.00
Batch output / 1MBatch price per 1M output tokens.
$5.00
—
1K chats2K tokens in, 500 out, each.
$9.00 best
$18.00
1K documents20K tokens in, 1K out, each.
$50.00 best
$100.00
1K agent tasks200K tokens in, 20K out, each.
$600.00 best
$1,200.00
ContextMaximum tokens in one request.
1.05M
1.05M
AgenticMulti-step tasks with tools.
72.9 best
71.6
CodingCode writing and repair.
78.6 best
70.2
ReasoningLogic problems and puzzles.
78.8 best
71.4
MultimodalTasks with images and text.
73.2
—
KnowledgeFacts and expert knowledge.
75.7 best
67.7
MultilingualTasks in many languages.
71.9
—
InstructionTasks with strict prompt rules.
—
72.8
MathMath problems.
—
72.1

Benchmarks

22 shared
Only GPT-6.2 Sol: 26Only Muse Wonder: 42
GPT-6.2 Sol
Muse Wonder
61.6% best7 / 170
38.0%62 / 170
87.6% best16 / 130
83.1%27 / 130
37.0%52 / 98
50.5% best10 / 98
24.8%49 / 89
36.3% best18 / 89
60.6% best7 / 68
49.5%15 / 68
13.4 best10 / 59
1.526 / 59
10.9 best12 / 59
2.526 / 59
18.0% best3 / 55
0.0%29 / 55
61.5% best7 / 52
47.0%17 / 52
58.6% best4 / 48
22.9%14 / 48
12.1%21 / 42
23.8% best4 / 42
77.8% best9 / 34
67.0%23 / 34
36.9% best11 / 34
27.5%19 / 34
43.8% best6 / 32
25.3%19 / 32
54.7% best5 / 30
31.7%18 / 30
71.3% best10 / 24
39.1%23 / 24
37.2% best6 / 24
22.2%13 / 24
30.0%4 / 23
30.6% best3 / 23
60.1% best6 / 23
52.0%12 / 23
65.9% best3 / 21
60.5%12 / 21