Claude Opus 5.5 vs GPT-6.2 Astra

Change models
IndexAll categories, out of 100.
77.1 ±2.3
79.5 ±3.1 best
GapIndex points below the top.
−2.4 · no clear gap
Top
CoverageIndex weight that has results.
95%
95%
SpeedOutput tokens per second.
84/s best
39/s
EloLMArena rating and rank.
1513 best
1414
ReleasedMonth of release.
2026-09
2026-11
Input / 1MUS dollars per 1M input tokens.
$4.00 best
$8.00
Output / 1MUS dollars per 1M output tokens.
$20.00 best
$40.00
Batch output / 1MBatch price per 1M output tokens.
$10.00 best
$20.00
1K chats2K tokens in, 500 out, each.
$18.00 best
$36.00
1K documents20K tokens in, 1K out, each.
$100.00 best
$200.00
1K agent tasks200K tokens in, 20K out, each.
$1,200.00 best
$2,400.00
ContextMaximum tokens in one request.
1M
1.05M best
AgenticMulti-step tasks with tools.
72.5
74.7 best
CodingCode writing and repair.
79.9 best
79.4
ReasoningLogic problems and puzzles.
79.0
86.2 best
MultimodalTasks with images and text.
74.6 best
70.7
KnowledgeFacts and expert knowledge.
78.8
79.9 best
MultilingualTasks in many languages.
—
74.0
InstructionTasks with strict prompt rules.
58.9
—
MathMath problems.
76.1
76.4 best

Benchmarks

50 shared
Only Claude Opus 5.5: 111Only GPT-6.2 Astra: 13
Claude Opus 5.5
GPT-6.2 Astra
92.3%31 / 220
96.8% best1 / 220
100.0%1 / 203
100.0%1 / 203
65.5%4 / 170
68.6% best2 / 170
91.7%8 / 137
95.8% best1 / 137
97.5%7 / 132
98.5% best1 / 132
90.3% best7 / 130
89.9%9 / 130
97.9% best3 / 127
97.5%6 / 127
98.3%10 / 116
99.3% best2 / 116
87.7%2 / 109
89.5% best1 / 109
50.5% best10 / 98
42.8%31 / 98
72.2%9 / 94
78.5% best2 / 94
Chess PuzzlesReasoning
67.0%7 / 93
78.0% best3 / 93
91.2%5 / 93
92.6% best4 / 93
44.5% best8 / 89
21.8%54 / 89
71.0%8 / 83
95.0% best1 / 83
95.0%7 / 75
100.0% best1 / 75
65.2% best1 / 68
63.1%4 / 68
18.0 best3 / 59
14.07 / 59
19.2 best2 / 59
12.410 / 59
13.0%11 / 55
17.0% best4 / 55