Inference
EN

DeepSeek V4 Pro 0813 vs Muse Spark Open

Change models
DeepSeek V4 Pro 0813DeepSeek
Muse Spark OpenMeta
IndexAll categories, out of 100.
65.5 ±3.7 best
65.1 ±3.9
GapIndex points below the top.
Top
−0.4 · no clear gap
CoverageIndex weight that has results.
85%
90% best
SpeedOutput tokens per second.
42/s
86/s best
EloLMArena rating and rank.
—
—
ReleasedMonth of release.
2026-08
2026-10
Input / 1MUS dollars per 1M input tokens.
$0.111 best
$1.25
Output / 1MUS dollars per 1M output tokens.
$4.44
$4.25 best
Batch output / 1MBatch price per 1M output tokens.
$1.98
—
1K chats2K tokens in, 500 out, each.
$2.44 best
$4.63
1K documents20K tokens in, 1K out, each.
$6.66 best
$29.25
1K agent tasks200K tokens in, 20K out, each.
$111.00 best
$335.00
ContextMaximum tokens in one request.
1.05M
1.05M
AgenticMulti-step tasks with tools.
63.7
64.8 best
CodingCode writing and repair.
65.1 best
63.3
ReasoningLogic problems and puzzles.
68.5 best
62.3
MultimodalTasks with images and text.
—
69.8
KnowledgeFacts and expert knowledge.
65.0
65.4 best
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
62.0
—
MathMath problems.
64.1
64.9 best

Benchmarks

30 shared
Only DeepSeek V4 Pro 0813: 57Only Muse Spark Open: 15
DeepSeek V4 Pro 0813
Muse Spark Open
Artificial Analysis Humanity's Last ExamKnowledge
41.0%43 / 222
43.9% best30 / 222
Artificial Analysis Long Context ReasoningReasoning
80.3% best40 / 217
78.3%63 / 217
GPQA diamond · Epoch AIKnowledge
91.7% best30 / 217
88.6%52 / 217
Critical Physics TasksReasoning
18.0% best35 / 214
14.9%51 / 214
OTIS Mock AIME 2024-2025Math
98.6% best27 / 200
98.1%31 / 200
Artificial Analysis GPQA DiamondKnowledge
92.8% best14 / 199
89.0%48 / 199
Artificial Analysis Omniscience AccuracyKnowledge
49.1% best22 / 164
42.0%44 / 164
GDPval-AA normalizedAgentic
54.5% best23 / 133
45.1%39 / 133
ARC-AGI-2 (semi-private)Reasoning
61.3% best38 / 131
57.5%46 / 131
ARC-AGI-1 (semi-private)Reasoning
90.0% best36 / 126
89.0%39 / 126
Vibe Code Bench v1.1Coding
82.3% best25 / 123
76.2%41 / 123
ARC-AGI-2 (public eval)Reasoning
59.4% best42 / 122
57.2%44 / 122
ARC-AGI-1 (public eval)Reasoning
93.3%39 / 116
94.0% best38 / 116
Artificial Analysis SciCodeCoding
51.0%48 / 114
58.4% best17 / 114
SimpleQA VerifiedKnowledge
52.9%26 / 91
56.6% best22 / 91
FrontierMath-Tiers-1-3-v2-PrivateMath
64.6%39 / 90
69.8% best30 / 90
Legal Research BenchLegal
40.9% best33 / 90
38.5%39 / 90
Tax Agent BenchFinance
26.8% best40 / 82
26.3%42 / 82
LiveBench Data AnalysisReasoning
79.2% best19 / 80
60.2%72 / 80
LiveBench ReasoningReasoning
85.8% best35 / 80
73.7%71 / 80