GPT-5.2 vs Muse Spark

Change models
GPT-5.2OpenAI
IndexAll categories, out of 100.
58.3 ±2.2
59.0 ±3.5 best
GapIndex points below the top.
−0.7 · no clear gap
Top
CoverageIndex weight that has results.
100% best
95%
SpeedOutput tokens per second.
54/s
—
EloLMArena rating and rank.
1412
1474 best
ReleasedMonth of release.
2025-12
2026-04
Input / 1MUS dollars per 1M input tokens.
$1.75
—
Output / 1MUS dollars per 1M output tokens.
$14.00
—
Batch output / 1MBatch price per 1M output tokens.
$7.00
—
1K chats2K tokens in, 500 out, each.
$10.50
—
1K documents20K tokens in, 1K out, each.
$49.00
—
1K agent tasks200K tokens in, 20K out, each.
$630.00
—
ContextMaximum tokens in one request.
400K best
262K
AgenticMulti-step tasks with tools.
55.6 best
55.1
CodingCode writing and repair.
59.7 best
52.7
ReasoningLogic problems and puzzles.
62.6 best
58.7
MultimodalTasks with images and text.
50.5
62.0 best
KnowledgeFacts and expert knowledge.
60.0
66.8 best
MultilingualTasks in many languages.
51.6
—
InstructionTasks with strict prompt rules.
54.8
59.0 best
MathMath problems.
61.7 best
59.5

Benchmarks

32 shared
Only GPT-5.2: 68Only Muse Spark: 15
GPT-5.2
Muse Spark
91.4% best36 / 220
89.8%46 / 220
37.7%56 / 216
40.7% best45 / 216
82.7% best23 / 211
78.0%70 / 211
11.6% best58 / 207
11.3%59 / 207
96.1% best40 / 203
88.9%66 / 203
90.3% best34 / 193
88.4%49 / 193
44.3%43 / 170
49.6% best25 / 170
82.8%65 / 149
84.2% best37 / 149
TaxEval v2Finance
75.8%11 / 143
77.7% best3 / 143
84.8%55 / 143
91.5% best36 / 143
75.4%22 / 140
75.9% best15 / 140
91.7% best18 / 135
89.6%28 / 135
MMLU ProKnowledge
86.2%45 / 135
87.3% best29 / 135
CorpFin v2Finance
65.9% best25 / 131
65.1%32 / 131
53.5% best70 / 130
19.7%99 / 130
80.0% best22 / 114
77.4%36 / 114
MedScribeHealthcare
84.4%38 / 111
85.9% best27 / 111
MedCodeHealthcare
49.7%20 / 108
51.3% best16 / 108
AIMEMath
96.9%2 / 95
96.9%2 / 95
MMMU ProMultimodal
86.7%17 / 92
87.4% best16 / 92