Enter opens the highlighted result. Esc closes.

DeepSeek V4 Pro 0423 vs Muse Spark 1.3

Muse Spark 1.3 leads DeepSeek V4 Pro 0423 on coding, 57.4 to 38.7 on the Vibe Coding Index, and DeepSeek V4 Pro 0423 costs 1.6x less per output token ($2.716 vs $4.25 per million). Updated October 21, 2026; 64 benchmarks have a result for both.

Replace DeepSeek V4 Pro 0423 Replace Muse Spark 1.3
DeepSeek V4 Pro 0423DeepSeek
Coding
38.7
Overall
50.0
Output $/M
$2.716
1
Muse Spark 1.3Meta
Coding
57.4
Overall
68.6
Output $/M
$4.25
63
vs boards won of 64 shared

DeepSeek V4 Pro 0423 or Muse Spark 1.3: which to pick

Pick Muse Spark 1.3 for coding: it scores 57.4 on the Vibe Coding Index against 38.7 for DeepSeek V4 Pro 0423, though on 4 benchmarks against 5.

Pick DeepSeek V4 Pro 0423 for output tokens 1.6x cheaper ($2.716 vs $4.25 per million) and work measured by Terminal-Bench 4 (Vals AI), where it leads.

Model ADeepSeek V4 Pro 0423DeepSeek38.7/1005 independent benchmarks$0.57 per solved taskModel BMuse Spark 1.3Meta57.4/1004 independent benchmarksNo cost per solved task
Shared results
64
DeepSeek V4 Pro 0423 only
20
Muse Spark 1.3 only
28
Like-for-like categories
8 of 9

Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.

Which one for your work

Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.

At a glance

Coding index
DeepSeek V4 Pro 0423 38.7 from 5 benchmarks, Muse Spark 1.3 57.4 from 4 benchmarks.
Widest category
Agentic work, 23.4 points apart on average over 8 shared benchmarks.
Widest single gap
FrontierMath, tier 4: DeepSeek V4 Pro 0423 2.4% against Muse Spark 1.3 46.3%, 43.9 points apart on a 0 to 100 scale.
Output price
DeepSeek V4 Pro 0423 $2.716 against Muse Spark 1.3 $4.25 per million output tokens, 1.6x apart.
Context window
Both take 1.05M tokens.

Benchmark by benchmark

First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.

Boards won, where both have a result
CategoryDeepSeek V4 Pro 0423BarsMuse Spark 1.3Ties
Coding111
Intelligence02
Agentic work08
Reasoning05
Math06
Knowledge04
Web design08
General019
All shared boards163
CodingLike-for-likeSix of these boards feed the Vibe Coding Index111
Coding benchmarks for DeepSeek V4 Pro 0423 and Muse Spark 1.3
BenchmarkDeepSeek V4 Pro 0423Muse Spark 1.3Leader
LiveBench codingNot measured90.2%Only Muse Spark 1.3 measured
LMArena WebDev14471659Muse Spark 1.3
AA Coding Index59.475.8Muse Spark 1.3
LiveCodeBench (Vals AI)87.5%Not measuredOnly DeepSeek V4 Pro 0423 measured
SWE-bench Verified (Vals AI)77.4%Not measuredOnly DeepSeek V4 Pro 0423 measured
Terminal-Bench 4 (Vals AI)11.1%10.6%DeepSeek V4 Pro 0423
Vibe Code Bench (Vals AI)49.9%82.9%Muse Spark 1.3
Code migration (Vals AI)26.2%27.6%Muse Spark 1.3
ProgramBench (Vals AI)0.0%0.5%Muse Spark 1.3
IOI (Vals AI)Not measured43.9%Only Muse Spark 1.3 measured
Terminal-Bench 2.1 (Vals AI)50.2%72.3%Muse Spark 1.3
SkillsBench (Vals AI)51.3%Not measuredOnly DeepSeek V4 Pro 0423 measured
CyberBench (Vals AI)Not measured69.4%Only Muse Spark 1.3 measured
SciCodeNot measured59.7%Only Muse Spark 1.3 measured
FrontierCode17.6%Not measuredOnly DeepSeek V4 Pro 0423 measured
FrontierSWENot measured25.8%Only Muse Spark 1.3 measured
WeirdML48.9%Not measuredOnly DeepSeek V4 Pro 0423 measured
ALE-Bench1006.11688.3Muse Spark 1.3
LMArena Text: coding14701510Muse Spark 1.3
LMArena Text: software and IT14691511Muse Spark 1.3
LMArena WebDev: React14411672Muse Spark 1.3
LMArena WebDev: HTML14351653Muse Spark 1.3
LMArena WebDev: from an imageNot measured1640Only Muse Spark 1.3 measured
Neither model has a result on the other 16 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for DeepSeek V4 Pro 0423 and Muse Spark 1.3
BenchmarkDeepSeek V4 Pro 0423Muse Spark 1.3Leader
AA Intelligence Index30.448.1Muse Spark 1.3
Epoch Capabilities Index149.1157.2Muse Spark 1.3
Agentic workLike-for-likePublic boards, shown as published, kept out of the index08
Agentic work benchmarks for DeepSeek V4 Pro 0423 and Muse Spark 1.3
BenchmarkDeepSeek V4 Pro 0423Muse Spark 1.3Leader
AA Agentic Index26.355.5Muse Spark 1.3
Design Arena: agents building web apps10001297Muse Spark 1.3
Design Arena: agents building full-stack apps9481295Muse Spark 1.3
Design Arena: agents building mobile appsNot measured1217Only Muse Spark 1.3 measured
Design Arena: agents building Godot games10591241Muse Spark 1.3
Design Arena: PowerPoint slidesNot measured1243Only Muse Spark 1.3 measured
Design Arena: spreadsheetsNot measured1043Only Muse Spark 1.3 measured
LiveBench researchNot measured68.6%Only Muse Spark 1.3 measured
LiveBench browserNot measured80.0%Only Muse Spark 1.3 measured
LiveBench long agenticNot measured40.4%Only Muse Spark 1.3 measured
Finance Agent (Vals AI)44.1%58.9%Muse Spark 1.3
Tax Agent Bench (Vals AI)25.2%36.0%Muse Spark 1.3
Legal Research Bench (Vals AI)23.1%40.9%Muse Spark 1.3
Excel Modeling Benchmark (Vals AI)51.6%62.7%Muse Spark 1.3
Terminal-Bench-Science (Vals AI)Not measured4.3%Only Muse Spark 1.3 measured
APEX-AgentsNot measured58.6%Only Muse Spark 1.3 measured
Vending-Bench 2$3,285Not measuredOnly DeepSeek V4 Pro 0423 measured
Neither model has a result on the other 24 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index05
Reasoning benchmarks for DeepSeek V4 Pro 0423 and Muse Spark 1.3
BenchmarkDeepSeek V4 Pro 0423Muse Spark 1.3Leader
GPQA Diamond89.4%Not measuredOnly DeepSeek V4 Pro 0423 measured
LiveBench reasoningNot measured89.7%Only Muse Spark 1.3 measured
GPQA Diamond (Artificial Analysis run)Not measured94.1%Only Muse Spark 1.3 measured
Humanity's Last Exam (Artificial Analysis run)Not measured48.7%Only Muse Spark 1.3 measured
GPQA Diamond (Epoch AI run)90.9%Not measuredOnly DeepSeek V4 Pro 0423 measured
Chess puzzles19.0%37.0%Muse Spark 1.3
Mystery game puzzles17.0%25.0%Muse Spark 1.3
LMArena Text: hard prompts14611503Muse Spark 1.3
LMArena Text: expert14631519Muse Spark 1.3
LMArena Text: hard prompts in English14661500Muse Spark 1.3
LMArena Vision: diagramsNot measured1330Only Muse Spark 1.3 measured
LMArena Vision: homeworkNot measured1296Only Muse Spark 1.3 measured
Kagi LLM Benchmark53.5%Not measuredOnly DeepSeek V4 Pro 0423 measured
Neither model has a result on the other 9 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index06
Math benchmarks for DeepSeek V4 Pro 0423 and Muse Spark 1.3
BenchmarkDeepSeek V4 Pro 0423Muse Spark 1.3Leader
OTIS Mock AIME96.7%99.2%Muse Spark 1.3
FrontierMath, tiers 1 to 345.3%74.4%Muse Spark 1.3
FrontierMath, tier 42.4%46.3%Muse Spark 1.3
ProofBench16.0%58.0%Muse Spark 1.3
LMArena Text: math14431497Muse Spark 1.3
LMArena Text: mathematical work14551491Muse Spark 1.3
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index04
Knowledge benchmarks for DeepSeek V4 Pro 0423 and Muse Spark 1.3
BenchmarkDeepSeek V4 Pro 0423Muse Spark 1.3Leader
MMLU Pro87.3%Not measuredOnly DeepSeek V4 Pro 0423 measured
LegalBench (Vals AI)80.3%Not measuredOnly DeepSeek V4 Pro 0423 measured
CaseLaw (Vals AI)59.4%Not measuredOnly DeepSeek V4 Pro 0423 measured
CorpFin (Vals AI)61.4%Not measuredOnly DeepSeek V4 Pro 0423 measured
TaxEval (Vals AI)72.1%Not measuredOnly DeepSeek V4 Pro 0423 measured
MedCode (Vals AI)40.5%Not measuredOnly DeepSeek V4 Pro 0423 measured
SimpleQA Verified47.0%Not measuredOnly DeepSeek V4 Pro 0423 measured
LMArena Text: science14701507Muse Spark 1.3
LMArena Text: medicine and healthcare14661493Muse Spark 1.3
LMArena Text: legal and government14641499Muse Spark 1.3
LMArena Text: business and finance14431490Muse Spark 1.3
Neither model has a result on the other 7 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index08
Web design benchmarks for DeepSeek V4 Pro 0423 and Muse Spark 1.3
BenchmarkDeepSeek V4 Pro 0423Muse Spark 1.3Leader
Design Arena: websites12341356Muse Spark 1.3
Design Arena: UI components12221351Muse Spark 1.3
Design Arena: game dev12291329Muse Spark 1.3
Design Arena: data viz12031343Muse Spark 1.3
Design Arena: 3D12541359Muse Spark 1.3
Design Arena: all code categories12401353Muse Spark 1.3
Design Arena: SVG11401301Muse Spark 1.3
Design Arena: ASCII art11491280Muse Spark 1.3
LiveBench designNot measured74.8%Only Muse Spark 1.3 measured
GeneralLike-for-likePublic boards, shown as published, kept out of the index019
General benchmarks for DeepSeek V4 Pro 0423 and Muse Spark 1.3
BenchmarkDeepSeek V4 Pro 0423Muse Spark 1.3Leader
LiveBench data analysisNot measured79.6%Only Muse Spark 1.3 measured
Public Benefits Bench (Vals AI)62.9%Not measuredOnly DeepSeek V4 Pro 0423 measured
MedScribe (Vals AI)75.1%Not measuredOnly DeepSeek V4 Pro 0423 measured
Long context reasoning (AA-LCR)Not measured83.0%Only Muse Spark 1.3 measured
DTBench90.7%96.5%Muse Spark 1.3
LMCA41.2%53.9%Muse Spark 1.3
Surface Evolver Bench40.0%Not measuredOnly DeepSeek V4 Pro 0423 measured
CL-bench Life13.5%Not measuredOnly DeepSeek V4 Pro 0423 measured
LMArena Text14511490Muse Spark 1.3
LMArena Text: instruction following14451479Muse Spark 1.3
LMArena Text: creative writing14461462Muse Spark 1.3
LMArena VisionNot measured1308Only Muse Spark 1.3 measured
LMArena Text: English14591495Muse Spark 1.3
LMArena Text: non-English14391481Muse Spark 1.3
LMArena Text: Chinese14811532Muse Spark 1.3
LMArena Text: German14581513Muse Spark 1.3
LMArena Text: Japanese14461470Muse Spark 1.3
LMArena Text: Korean14091490Muse Spark 1.3
LMArena Text: French14731511Muse Spark 1.3
LMArena Text: Spanish14571496Muse Spark 1.3
LMArena Text: Russian14531491Muse Spark 1.3
LMArena Text: Polish14611490Muse Spark 1.3
LMArena Text: multi-turn14681487Muse Spark 1.3
LMArena Text: longer queries14581488Muse Spark 1.3
LMArena Text: writing and language14471470Muse Spark 1.3
LMArena Text: entertainment and media14311457Muse Spark 1.3
LMArena Vision: EnglishNot measured1298Only Muse Spark 1.3 measured
LMArena Vision: ChineseNot measured1384Only Muse Spark 1.3 measured
LMArena Vision: reading textNot measured1319Only Muse Spark 1.3 measured
LMArena Vision: humorNot measured1293Only Muse Spark 1.3 measured
LMArena Vision: creative writingNot measured1296Only Muse Spark 1.3 measured
Neither model has a result on the other 15 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for DeepSeek V4 Pro 0423 and Muse Spark 1.3
BenchmarkDeepSeek V4 Pro 0423Muse Spark 1.3Leader
Neither model has a result on 5 boards here

What each workload costs

Three fixed token mixes at each model's listed price, so per-token rates become a bill.

Chat turn

1K input + 500 output tokens

DeepSeek V4 Pro 0423$0.0026Fits in one request
Muse Spark 1.3$0.0034Fits in one request

DeepSeek V4 Pro 0423 costs less

Repository review

50K input + 3K output tokens

DeepSeek V4 Pro 0423$0.069Fits in one request
Muse Spark 1.3$0.075Fits in one request

DeepSeek V4 Pro 0423 costs less

Agent loop

200K input + 10K output tokens

DeepSeek V4 Pro 0423$0.272Fits in one request
Muse Spark 1.3$0.292Fits in one request

DeepSeek V4 Pro 0423 costs less

Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.

Specification differences

The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.

Specifications of DeepSeek V4 Pro 0423 and Muse Spark 1.3
SpecificationDeepSeek V4 Pro 0423Meta Muse Spark 1.3
MakerDeepSeekMeta
WeightsOpen weightsNo public weights listed
ReleasedApril 24, 2026September 2, 2026
Context window1.05M1.05M
Input $/Mtok$1.225$1.25
Output $/Mtok$2.716$4.25
Coding index38.757.4
Independent measurements behind it54
Cost per solved task$0.57Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is DeepSeek V4 Pro 0423 better than Muse Spark 1.3 for coding?

Pick Muse Spark 1.3 for coding: it scores 57.4 on the Vibe Coding Index against 38.7 for DeepSeek V4 Pro 0423, though on 4 benchmarks against 5. Pick DeepSeek V4 Pro 0423 for output tokens 1.6x cheaper ($2.716 vs $4.25 per million) and work measured by Terminal-Bench 4 (Vals AI), where it leads.

Which is cheaper, DeepSeek V4 Pro 0423 or Muse Spark 1.3?

DeepSeek V4 Pro 0423 costs $1.225 per million input tokens and $2.716 per million output tokens; Muse Spark 1.3 costs $1.25 and $4.25. Cost per solved task is known for DeepSeek V4 Pro 0423 only.

Which has the bigger context window, DeepSeek V4 Pro 0423 or Muse Spark 1.3?

Neither: both take 1.05M tokens.

How many benchmarks is the DeepSeek V4 Pro 0423 vs Muse Spark 1.3 comparison based on?

64 benchmarks have a result for both models as of October 21, 2026. DeepSeek V4 Pro 0423's coding index rests on 5 independent benchmarks and Muse Spark 1.3's on 4. Where only one model has a result, the page says so instead of naming a winner.

Compare next

Embed this comparison on your site

Free under CC BY 4.0, while the source link stays visible. Preview the frame · Terms and other embeds

<iframe src="https://llmbenchmark.ai/embed/compare/deepseek-v4-pro-vs-meta-muse-spark-1-3" title="DeepSeek V4 Pro 0423 vs Muse Spark 1.3, LLMBenchmark.ai" width="100%" height="360" style="border:0;max-width:100%" loading="lazy"></iframe>
<p style="font:13px/1.4 sans-serif;margin:6px 0 0">Source: <a href="https://llmbenchmark.ai/compare/deepseek-v4-pro-vs-meta-muse-spark-1-3">LLMBenchmark.ai, DeepSeek V4 Pro 0423 vs Muse Spark 1.3</a></p>