GPT-6.2 Astra vs Muse Wonder Full

Change models
IndexAll categories, out of 100.
79.5 ±3.1 best
75.8 ±5.3
GapIndex points below the top.
Top
−3.7 · no clear gap
CoverageIndex weight that has results.
95% best
80%
SpeedOutput tokens per second.
39/s
—
EloLMArena rating and rank.
1414
—
ReleasedMonth of release.
2026-11
2026-10
Input / 1MUS dollars per 1M input tokens.
$8.00
—
Output / 1MUS dollars per 1M output tokens.
$40.00
—
Batch output / 1MBatch price per 1M output tokens.
$20.00
—
1K chats2K tokens in, 500 out, each.
$36.00
—
1K documents20K tokens in, 1K out, each.
$200.00
—
1K agent tasks200K tokens in, 20K out, each.
$2,400.00
—
ContextMaximum tokens in one request.
1.05M best
1M
AgenticMulti-step tasks with tools.
74.7
76.0 best
CodingCode writing and repair.
79.4 best
75.9
ReasoningLogic problems and puzzles.
86.2 best
75.2
MultimodalTasks with images and text.
70.7
—
KnowledgeFacts and expert knowledge.
79.9 best
70.7
MultilingualTasks in many languages.
74.0
—
InstructionTasks with strict prompt rules.
—
—
MathMath problems.
76.4
78.3 best

Benchmarks

19 shared
Only GPT-6.2 Astra: 44Only Muse Wonder Full: 24
GPT-6.2 Astra
Muse Wonder Full
68.6% best2 / 170
41.0%50 / 170
89.9% best9 / 130
86.0%19 / 130
42.8%31 / 98
53.4% best9 / 98
21.8%54 / 89
43.0% best10 / 89
63.1% best4 / 68
58.1%9 / 68
14.0 best7 / 59
7.517 / 59
12.4 best10 / 59
10.413 / 59
17.0%4 / 55
17.0%4 / 55
71.4% best1 / 48
37.1%11 / 48
16.3%12 / 42
23.3% best5 / 42
80.7% best3 / 34
73.3%12 / 34
53.2% best1 / 34
35.1%14 / 34
62.7% best2 / 30
42.0%10 / 30
77.2% best4 / 24
45.4%22 / 24
39.4% best4 / 24
22.2%13 / 24
29.8%5 / 23
32.6% best1 / 23
64.1% best3 / 23
57.1%9 / 23
66.3% best2 / 21
62.3%9 / 21
15.0%
13.1%