GPT-5.4 Pro vs GPT-5.5

Change models
GPT-5.5OpenAI
IndexAll categories, out of 100.
73.3 ±8.7 best
69.1 ±2.5
GapIndex points below the top.
Top
−4.2 · no clear gap
CoverageIndex weight that has results.
60%
95% best
SpeedOutput tokens per second.
21/s
52/s best
EloLMArena rating and rank.
—
1466
ReleasedMonth of release.
2026-03
2026-04
Input / 1MUS dollars per 1M input tokens.
$30.00
$5.00 best
Output / 1MUS dollars per 1M output tokens.
$180.00
$30.00 best
Batch output / 1MBatch price per 1M output tokens.
$90.00
$15.00 best
1K chats2K tokens in, 500 out, each.
$150.00
$25.00 best
1K documents20K tokens in, 1K out, each.
$780.00
$130.00 best
1K agent tasks200K tokens in, 20K out, each.
$9,600.00
$1,600.00 best
ContextMaximum tokens in one request.
1.05M
1.05M
AgenticMulti-step tasks with tools.
70.5 best
65.8
CodingCode writing and repair.
—
65.5
ReasoningLogic problems and puzzles.
80.4 best
74.1
MultimodalTasks with images and text.
—
66.8
KnowledgeFacts and expert knowledge.
67.2
70.4 best
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
—
64.8
MathMath problems.
75.3 best
74.7

Benchmarks

15 shared
Only GPT-5.4 Pro: 3Only GPT-5.5: 97
GPT-5.4 Pro
GPT-5.5
94.6% best8 / 217
94.0%13 / 217
30.0% best10 / 214
27.1%17 / 214
83.3%19 / 131
85.0% best13 / 131
94.5%21 / 126
95.0% best19 / 126
92.2% best12 / 122
90.6%14 / 122
98.3% best8 / 116
97.5%19 / 116
46.3%40 / 91
63.0% best16 / 91
Chess PuzzlesReasoning
59.0%10 / 90
63.0% best7 / 90
82.5%17 / 90
85.3% best16 / 90
58.7% best8 / 77
52.2%19 / 77
58.5%20 / 72
72.5% best16 / 72
50.0%3 / 69
51.7% best1 / 69
BrowseCompAgentic
89.3% best11 / 63
84.4%19 / 63
37.5% best3 / 55
35.4%4 / 55
42.7% best25 / 53
41.4%26 / 53