DeepSeek V4.1 Flash 1017 vs Muse Spark 1.3

Change models
IndexAll categories, out of 100.
68.5 ±2.6
70.1 ±4.0 best
GapIndex points below the top.
−1.6 · no clear gap
Top
CoverageIndex weight that has results.
95% best
85%
SpeedOutput tokens per second.
261/s best
76/s
EloLMArena rating and rank.
1463
1491 best
ReleasedMonth of release.
2026-10
2026-09
Input / 1MUS dollars per 1M input tokens.
$0.022 best
$1.25
Output / 1MUS dollars per 1M output tokens.
$0.40 best
$4.25
Batch output / 1MBatch price per 1M output tokens.
$0.20
—
1K chats2K tokens in, 500 out, each.
$0.244 best
$4.63
1K documents20K tokens in, 1K out, each.
$0.84 best
$29.25
1K agent tasks200K tokens in, 20K out, each.
$12.40 best
$335.00
ContextMaximum tokens in one request.
1.05M
1.05M
AgenticMulti-step tasks with tools.
72.2 best
70.4
CodingCode writing and repair.
69.8 best
68.8
ReasoningLogic problems and puzzles.
65.0
68.1 best
MultimodalTasks with images and text.
62.2
—
KnowledgeFacts and expert knowledge.
64.2
67.5 best
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
68.3
84.4 best
MathMath problems.
65.4
68.6 best

Benchmarks

47 shared
Only DeepSeek V4.1 Flash 1017: 47Only Muse Spark 1.3: 12
DeepSeek V4.1 Flash 1017
Muse Spark 1.3
39.2%51 / 222
48.7% best15 / 222
84.0% best8 / 217
83.0%18 / 217
14.3%52 / 214
24.9% best21 / 214
98.3%28 / 200
99.2% best19 / 200
46.4% best33 / 164
43.6%41 / 164
55.0%21 / 133
59.0% best10 / 133
84.7% best18 / 123
82.9%23 / 123
51.9%44 / 114
58.8% best14 / 114
Chess PuzzlesReasoning
48.0% best19 / 90
37.0%39 / 90
67.4%33 / 90
74.0% best25 / 90
41.3% best30 / 90
40.9%33 / 90
90.6% best3 / 86
88.8%7 / 86
26.6%41 / 82
37.9% best14 / 82
79.3%15 / 80
79.6% best13 / 80
86.7%31 / 80
89.7% best15 / 80
43.0% best17 / 80
25.0%37 / 80
53.5%33 / 79
58.9% best8 / 79
45.6% best17 / 77
27.6%44 / 77
EMBFinance
57.2%40 / 75
62.7% best31 / 75
74.5% best18 / 75
72.3%24 / 75