Compare / DeepSeek-V4-Pro (0423) vs Muse Spark 1.3

DeepSeek-V4-Pro (0423) vs Muse Spark 1.3

Which is cheaper, which is more capable, and which does better at each kind of work, from 42 results on 4 sources that measured both models.

DeepSeekDeepSeek-V4-Pro (0423) Released 24 Apr 2026 · $1.42 in · $2.83 out
MetaMuse Spark 1.3 Released 2 Sep 2026 · $1.25 in · $4.25 out
DeepSeek-V4-Pro (0423) wins 27 too close to callMuse Spark 1.3 wins 33

Of the 42 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.

Results as of 16 November 2026 (catalogue release 2026-11-16-49bdfa045dee). See also DeepSeek vs Meta, every model from both labs.

DeepSeek-V4-Pro (0423) or Muse Spark 1.3? The short answer

Pick Muse Spark 1.3: it scores higher (Spring Prompt overall 59 against 28) at about the same price.

  • Spring Prompt overall: DeepSeek-V4-Pro (0423) 28, Muse Spark 1.3 59 (out of 100; how the score works).
  • They cost about the same to run: $0.046 and $0.046 per deck on DeckBench.
  • Muse Spark 1.3 scores higher on the Artificial Analysis Intelligence Index: 48.1 against 30.4.
  • Muse Spark 1.3 is newer: released 2 Sep 2026, against 24 Apr 2026 for DeepSeek-V4-Pro (0423).
  • On our own benchmarks, on the results both models have: Muse Spark 1.3 does better for decks from an analysis and marketing planning.
  • On third-party benchmarks, on the results both models have: Muse Spark 1.3 does better for agents and tool use, reasoning and knowledge, writing and sticking to the facts.

Head to head

28
Spring Prompt overall
59
30.4
Intelligence Index
48.1
$1.77
Price per million tokens, blendedlower is better
$2.00
1,048,576
Context window, tokens
1,048,576

Every fact side by side

DeepSeek-V4-Pro (0423)Muse Spark 1.3
Price per million tokens$1.42 in · $2.83 out$1.25 in · $4.25 out
Blended (3 in : 1 out)$1.77$2.00
Cheapest host, blended$1.05 (Relace, FP4)–
Cost per deck (DeckBench)$0.046$0.046
Cached input, per million–$0.15 cached input
Intelligence Index30.448.1
Spring Prompt overall2859
Output speed, tokens a second–254
Context window1,048,576 tokens1,048,576 tokens
Released24 Apr 20262 Sep 2026 (newer)
WeightsOpenClosed
Developer based inChinathe US

Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.

Which is better for what

Each model's rank among the models on sale today, for each kind of work, from our own benchmarks and licensed sources. Further right is better; the better rank is in green.

Product listingsTurning a sparse product feed and photos into listings that can go live not measured 17 of 32
Decks from an analysisTurning a finished analysis into a deck you could present as it is 32 of 32 25 of 32
Marketing planningPlanning a year of ad spend without overspending 26 of 32 25 of 32
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 92 of 159 13 of 159
Professional workReal tasks from banking, consulting and law, business documents and freelance projects not measured 8 of 51
Reasoning and knowledgeHard questions across science, maths and general knowledge 51 of 187 11 of 187
WritingWhat people prefer in blind comparisons, and judged writing quality 30 of 169 7 of 169
Sticking to the factsSummarising without inventing things, and factual answers 56 of 139 11 of 139
Speed under pressureGood decisions against a real clock (fast chess) 36 of 36 33 of 35

From our research

Every shared result

Each model's best result on every measure both have, grouped by benchmark, with the setting that produced it. Under each measure, every model that benchmark has measured is a grey tick, best to the right, so you can see where the two sit in the field. The headline measures of each benchmark come first; the rest are under "Show all". Green marks the better value; where a source reports ranges that overlap, neither is marked.

MeasureDeepSeek-V4-Pro (0423)Muse Spark 1.3
DeckBench measured by Spring Prompt · 3 results
Cost per deckUS dollars, lower is better 14 · 13 of 33 measured $0.0465 $0.0463
Deck ratingrating, higher is betterThe rating weighs each win by how strong the other deck was; the win rate counts wins against whichever decks a model happened to meet, so the two can disagree. 33 · 25 of 33 measured 542 819
Accurate decks% of tasks, higher is better 28 · 24 of 33 measured 0.0% 16.7%
ROASBench measured by Spring Prompt · 3 results
Overall scorescore out of 100, higher is better 26 · 25 of 33 measured 28.3 34.1
Audience scorescore out of 100, higher is better 29 · 25 of 33 measured 44.8 51.6
Business scorescore out of 100, higher is better 26 · 25 of 33 measured 23.8 25.8
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results
Overall · TextArena rating, higher is better 53 · 8 of 387 measured 1,458 1,494max reasoning
Overall · Text factualityArena rating, higher is better 49 · 13 of 163 measured 1,455 1,480max reasoning
Business, management and finance · TextArena rating, higher is better 52 · 4 of 380 measured 1,460 1,505max reasoning
Artificial Analysis reported by Artificial Analysis · 3 results
Artificial Analysis Intelligence Indexindex score, higher is better 53 · 10 of 249 measured 30.4max reasoning 48.1max reasoning
GPQA Diamond% of questions, higher is better 37 · 5 of 232 measured 90.5%high reasoning 94.1%xhigh reasoning
Humanity's Last Exam% of questions, higher is better 57 · 13 of 250 measured 37.5%max reasoning 48.7%max reasoning
Show all 42 results (30 more)
MeasureDeepSeek-V4-Pro (0423)Muse Spark 1.3
DeckBench measured by Spring Prompt · 14 results
Caveat dropped% of tasks, lower is better 15 · 15 of 33 measured 16.7% 16.7%
Clean layout% of tasks, higher is better 16 · 29 of 33 measured 33.3% 0.0%
Design quality% of the maximum, higher is better 31 · 25 of 33 measured 40.9% 56.1%
Draft figure quoted% of tasks, lower is better 19 · 13 of 33 measured 33.3% 16.7%
Findings missing% of tasks, lower is better 23 · 1 of 33 measured 16.7% 0.0%
Head-to-head win rate% of comparisons, higher is betterThe rating weighs each win by how strong the other deck was; the win rate counts wins against whichever decks a model happened to meet, so the two can disagree. 33 · 24 of 33 measured 24.4% 41.8%
Layout defects% of tasks, lower is better 16 · 29 of 33 measured 66.7% 100.0%
Misleading metric used% of tasks, lower is better 1 · 1 of 33 measured 0.0% 0.0%
Presentable decks% of tasks, higher is better 15 · 15 of 33 measured 0.0% 0.0%
Recommendation late or wrong% of tasks, lower is better 28 · 28 of 33 measured 66.7% 66.7%
Slides needing work% of tasks, lower is better 31 · 17 of 33 measured 83.3% 50.0%
Time per deckseconds, lower is better 28 · 12 of 33 measured 332 89.4
Unsupported claims% of tasks, lower is better 32 · 15 of 33 measured 66.7% 16.7%
Unsupported numbers% of tasks, lower is better 1 · 1 of 33 measured 0.0% 0.0%
ROASBench measured by Spring Prompt · 6 results
Consistency scorescore out of 100, higher is better 31 · 20 of 33 measured 9.1 26.7
Contribution profitsimulated US dollars, higher is better 25 · 24 of 33 measured $301,289 $337,251
Cost of a runUS dollars, lower is better 17 · 16 of 33 measured $0.20 $0.18
Months over budgetmonths of 12, lower is better 1 · 1 of 33 measured 0 0
Planning scorescore out of 100, higher is better 30 · 15 of 33 measured 54.0 54.8
Return on ad spendprofit per $1 spent, higher is better 25 · 24 of 33 measured $0.26 $0.29
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 4 results
Creative writing · TextArena rating, higher is better 39 · 22 of 385 measured 1,446 1,459max reasoning
Expert prompts · TextArena rating, higher is better 64 · 12 of 338 measured 1,480 1,529max reasoning
Instruction following · TextArena rating, higher is better 52 · 13 of 387 measured 1,452 1,486max reasoning
Writing, literature and language · TextArena rating, higher is better 44 · 19 of 386 measured 1,451 1,470max reasoning
Artificial Analysis reported by Artificial Analysis · 6 results
Terminal-Bench 2.1% of tasks, higher is better 42 · 11 of 131 measured 64.8%high reasoning 85.4%xhigh reasoning
Terminal-Bench 4.0% of tasks, higher is better 32 · 19 of 120 measured 14.6%max reasoning 33.3%max reasoning
τ-bench banking% of tasks, higher is better 35 · 3 of 121 measured 30.1%max reasoning 50.5%max reasoning
Artificial Analysis Coding Indexindex score, higher is better 40 · 7 of 130 measured 59.4max reasoning 76.5xhigh reasoning
Long-context reasoning (AA-LCR)% of questions, higher is better 86 · 16 of 241 measured 74.7%max reasoning 83.0%max reasoning
SciCode% of problems, higher is better 50 · 10 of 121 measured 50.8%max reasoning 59.7%xhigh reasoning

Quick answers

Is DeepSeek-V4-Pro (0423) better than Muse Spark 1.3?

Pick Muse Spark 1.3: it scores higher (Spring Prompt overall 59 against 28) at about the same price.

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

They cost about the same to run: $0.046 and $0.046 per deck on DeckBench.

Which is better for decks from an analysis, DeepSeek-V4-Pro (0423) or Muse Spark 1.3?

Muse Spark 1.3 ranks 25 of 32 models on sale for turning a finished analysis into a deck you could present as it is, against 32 for DeepSeek-V4-Pro (0423).

Which is better for marketing planning, DeepSeek-V4-Pro (0423) or Muse Spark 1.3?

Muse Spark 1.3 ranks 25 of 32 models on sale for planning a year of ad spend without overspending, against 26 for DeepSeek-V4-Pro (0423).

Which is better for agents and tool use, DeepSeek-V4-Pro (0423) or Muse Spark 1.3?

Muse Spark 1.3 ranks 13 of 159 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 92 for DeepSeek-V4-Pro (0423).

Which is better for reasoning and knowledge, DeepSeek-V4-Pro (0423) or Muse Spark 1.3?

Muse Spark 1.3 ranks 11 of 187 models on sale for hard questions across science, maths and general knowledge, against 51 for DeepSeek-V4-Pro (0423).

Which is better for writing, DeepSeek-V4-Pro (0423) or Muse Spark 1.3?

Muse Spark 1.3 ranks 7 of 169 models on sale for what people prefer in blind comparisons, and judged writing quality, against 30 for DeepSeek-V4-Pro (0423).

Run DeepSeek-V4-Pro (0423) and Muse Spark 1.3 on your own prompt.

Benchmarks aren't your task. Test both side by side on your prompt, with every answer checked against your rules and the cost of each one.