Muse Spark 1.1 vs Muse Spark 1.2

Change models
IndexAll categories, out of 100.
64.6 ±3.1
64.9 ±3.3 best
GapIndex points below the top.
−0.3 · no clear gap
Top
CoverageIndex weight that has results.
95%
95%
SpeedOutput tokens per second.
152/s best
123/s
EloLMArena rating and rank.
1479
1484 best
ReleasedMonth of release.
2026-07
2026-08
Input / 1MUS dollars per 1M input tokens.
$1.25
$1.25
Output / 1MUS dollars per 1M output tokens.
$4.25
$4.25
1K chats2K tokens in, 500 out, each.
$4.63
$4.63
1K documents20K tokens in, 1K out, each.
$29.25
$29.25
1K agent tasks200K tokens in, 20K out, each.
$335.00
$335.00
ContextMaximum tokens in one request.
1.05M
1.05M
AgenticMulti-step tasks with tools.
61.9
62.7 best
CodingCode writing and repair.
64.8 best
64.3
ReasoningLogic problems and puzzles.
63.0
64.1 best
MultimodalTasks with images and text.
66.5
67.2 best
KnowledgeFacts and expert knowledge.
67.4 best
65.2
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
67.5
77.0 best
MathMath problems.
61.2
64.4 best

Benchmarks

38 shared
Only Muse Spark 1.1: 21Only Muse Spark 1.2: 19
Muse Spark 1.1
Muse Spark 1.2
46.2% best24 / 222
45.5%27 / 222
77.7%74 / 217
79.0% best58 / 217
15.1%50 / 214
17.7% best37 / 214
89.8%38 / 199
90.4% best33 / 199
52.1% best14 / 164
45.4%36 / 164
85.0%27 / 149
85.3% best22 / 149
TaxEval v2Finance
79.7%2 / 143
80.4% best1 / 143
MMLU ProKnowledge
88.7% best17 / 135
88.3%20 / 135
35.6%50 / 133
48.6% best28 / 133
CorpFin v2Finance
71.3% best4 / 131
70.9%5 / 131
72.2%46 / 123
79.1% best32 / 123
58.8% best14 / 114
57.4%19 / 114
MedScribeHealthcare
88.9%9 / 109
90.1% best5 / 109
66.1% best37 / 97
65.4%43 / 97
MMMU ProMultimodal
86.6% best18 / 92
86.1%22 / 92
57.8%20 / 91
60.3% best18 / 91
Chess PuzzlesReasoning
34.0% best47 / 90
27.0%52 / 90
38.0%43 / 90
43.8% best25 / 90
SWE-benchCoding
82.0%21 / 87
86.6% best13 / 87
80.0%37 / 86
82.9% best26 / 86