GPT-6.1 Terra vs GPT-6.2 Astra

Change models
IndexAll categories, out of 100.
75.1 ±3.6
79.5 ±3.1 best
GapIndex points below the top.
−4.4 · no clear gap
Top
CoverageIndex weight that has results.
90%
95% best
SpeedOutput tokens per second.
50/s best
39/s
EloLMArena rating and rank.
1426 best
1414
ReleasedMonth of release.
2026-10
2026-11
Input / 1MUS dollars per 1M input tokens.
$1.00 best
$8.00
Output / 1MUS dollars per 1M output tokens.
$5.00 best
$40.00
Batch output / 1MBatch price per 1M output tokens.
$2.50 best
$20.00
1K chats2K tokens in, 500 out, each.
$4.50 best
$36.00
1K documents20K tokens in, 1K out, each.
$25.00 best
$200.00
1K agent tasks200K tokens in, 20K out, each.
$300.00 best
$2,400.00
ContextMaximum tokens in one request.
1.05M
1.05M
AgenticMulti-step tasks with tools.
76.1 best
74.7
CodingCode writing and repair.
72.8
79.4 best
ReasoningLogic problems and puzzles.
79.3
86.2 best
MultimodalTasks with images and text.
70.4
70.7 best
KnowledgeFacts and expert knowledge.
69.9
79.9 best
MultilingualTasks in many languages.
—
74.0
InstructionTasks with strict prompt rules.
—
—
MathMath problems.
75.2
76.4 best

Benchmarks

32 shared
Only GPT-6.1 Terra: 22Only GPT-6.2 Astra: 31
GPT-6.1 Terra
GPT-6.2 Astra
94.3%14 / 220
96.8% best1 / 220
100.0%1 / 203
100.0%1 / 203
47.6%32 / 170
68.6% best2 / 170
84.0%24 / 130
89.9% best9 / 130
84.7%7 / 109
89.5% best1 / 109
26.0%69 / 98
42.8% best31 / 98
57.3%24 / 94
78.5% best2 / 94
90.9%7 / 93
92.6% best4 / 93
17.1%64 / 89
21.8% best54 / 89
74.0%6 / 83
95.0% best1 / 83
92.7%8 / 75
100.0% best1 / 75
49.5%15 / 68
63.1% best4 / 68
13.211 / 59
14.0 best7 / 59
8.516 / 59
12.4 best10 / 59
16.0%8 / 55
17.0% best4 / 55
53.3%14 / 52
62.1% best6 / 52
1.7%42 / 42
16.3% best12 / 42
79.2%7 / 40
89.2% best2 / 40
76.7% best6 / 40
67.8%8 / 40
EBR-benchReasoning
78.1%4 / 33
92.4% best1 / 33