GPT-5.5 vs GPT-6.2 Astra

Change models
GPT-5.5OpenAI
IndexAll categories, out of 100.
67.6 ±2.4
79.5 ±3.1 best
GapIndex points below the top.
−11.9
Top
CoverageIndex weight that has results.
95%
95%
SpeedOutput tokens per second.
39/s
39/s
EloLMArena rating and rank.
1466 best
1414
ReleasedMonth of release.
2026-04
2026-11
Input / 1MUS dollars per 1M input tokens.
$5.00 best
$8.00
Output / 1MUS dollars per 1M output tokens.
$30.00 best
$40.00
Batch output / 1MBatch price per 1M output tokens.
$15.00 best
$20.00
1K chats2K tokens in, 500 out, each.
$25.00 best
$36.00
1K documents20K tokens in, 1K out, each.
$130.00 best
$200.00
1K agent tasks200K tokens in, 20K out, each.
$1,600.00 best
$2,400.00
ContextMaximum tokens in one request.
1.05M
1.05M
AgenticMulti-step tasks with tools.
64.3
74.7 best
CodingCode writing and repair.
64.0
79.4 best
ReasoningLogic problems and puzzles.
72.3
86.2 best
MultimodalTasks with images and text.
65.9
70.7 best
KnowledgeFacts and expert knowledge.
69.3
79.9 best
MultilingualTasks in many languages.
—
74.0
InstructionTasks with strict prompt rules.
63.7
—
MathMath problems.
73.2
76.4 best

Benchmarks

29 shared
Only GPT-5.5: 86Only GPT-6.2 Astra: 34
GPT-5.5
GPT-6.2 Astra
94.0%17 / 220
96.8% best1 / 220
100.0%1 / 203
100.0%1 / 203
58.0%13 / 170
68.6% best2 / 170
93.6%10 / 144
97.0% best1 / 144
85.0%16 / 137
95.8% best1 / 137
95.0%22 / 132
98.5% best1 / 132
69.8%57 / 130
89.9% best9 / 130
90.6%16 / 127
97.5% best6 / 127
97.5%21 / 116
99.3% best2 / 116
79.9%24 / 109
89.5% best1 / 109
40.4%40 / 98
42.8% best31 / 98
63.0%19 / 94
78.5% best2 / 94
Chess PuzzlesReasoning
63.0%9 / 93
78.0% best3 / 93
85.3%19 / 93
92.6% best4 / 93
21.8%54 / 89
21.8%54 / 89
93.6%10 / 87
97.0% best1 / 87
56.0%15 / 83
95.0% best1 / 83
72.5%19 / 75
100.0% best1 / 75
0.0%29 / 55
17.0% best4 / 55
44.2%19 / 40
89.2% best2 / 40