DeepSeek V4 Pro 0813 vs Muse Spark 1.2
Updated October 21, 2026. 63 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.
Decision reading
Pick Muse Spark 1.2 for coding: it scores 65.1 on the Vibe Coding Index against 54.6 for DeepSeek V4 Pro 0813.
Pick DeepSeek V4 Pro 0813 when output tokens cost 2.1x less, or when a solved task costs $0.10 against $0.64.
- Shared results
- 63
- DeepSeek V4 Pro 0813 only
- 15
- Muse Spark 1.2 only
- 34
- Like-for-like categories
- 7 of 9
Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.
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.
Coding
Vibe Coding Index, four public coding benchmarks
Muse Spark 1.2
Muse Spark 1.2 scores 65.1 against 54.6.
1 and 2 benchmarks behind the scores
Intelligence
AA Intelligence Index
Muse Spark 1.2
Muse Spark 1.2 leads, 39.6 against 36.0.
Same board, both measured
Agentic work
AA Agentic Index
Muse Spark 1.2
Muse Spark 1.2 leads, 43.2 against 41.3.
Same board, both measured
Reasoning
GPQA Diamond
No call
Only DeepSeek V4 Pro 0813 has a GPQA Diamond result, 92.4%.
Directional only
Knowledge
MMLU Pro
Muse Spark 1.2
Muse Spark 1.2 leads, 88.3% against 87.0%.
Same board, both measured
Web design
Design Arena: websites
No call
Only Muse Spark 1.2 has a Design Arena: websites result, 1313.
Directional only
General
Vals Index
No call
Only DeepSeek V4 Pro 0813 has a Vals Index result, 53.0%.
Directional only
Long documents
Listed context window
No call
Both take 1.05M tokens.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
DeepSeek V4 Pro 0813
DeepSeek V4 Pro 0813 $0.0016, Muse Spark 1.2 $0.0034 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
DeepSeek V4 Pro 0813
DeepSeek V4 Pro 0813 $0.10, Muse Spark 1.2 $0.64.
Published attempt cost and pass rate
At a glance
- Coding index
- DeepSeek V4 Pro 0813 54.6 from 1 benchmark, Muse Spark 1.2 65.1 from 2 benchmarks.
- Widest category
- Coding, 7.8 points apart on average over 17 shared benchmarks.
- Widest single gap
- IOI (Vals AI): DeepSeek V4 Pro 0813 51.6% against Muse Spark 1.2 21.8%, 29.8 points apart on a 0 to 100 scale.
- Output price
- DeepSeek V4 Pro 0813 $1.98 against Muse Spark 1.2 $4.25 per million output tokens, 2.1x 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.
| Category | DeepSeek V4 Pro 0813 | Bars | Muse Spark 1.2 | Ties |
|---|---|---|---|---|
| Coding | 11 | 6 | ||
| Intelligence | 1 | 1 | ||
| Agentic work | 2 | 5 | ||
| Reasoning | 2 | 4 | ||
| Math | 1 | 2 | ||
| Knowledge | 0 | 10 | ||
| General | 2 | 16 | ||
| All shared boards | 19 | 44 |
CodingLike-for-likeFour of these boards feed the Vibe Coding Index116
| Benchmark | DeepSeek V4 Pro 0813 | Muse Spark 1.2 | Leader |
|---|---|---|---|
| LiveBench coding | Not measured | 78.2% | Only Muse Spark 1.2 measured |
| LMArena WebDev | 1582 | 1534 | DeepSeek V4 Pro 0813 |
| AA Coding Index | 68.8 | 72.2 | Muse Spark 1.2 |
| LiveCodeBench (Vals AI) | 87.5% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| SWE-bench Verified (Vals AI) | 96.4% | 86.6% | DeepSeek V4 Pro 0813 |
| Terminal-Bench 4 (Vals AI) | 14.1% | 6.1% | DeepSeek V4 Pro 0813 |
| Vibe Code Bench (Vals AI) | 82.3% | 79.1% | DeepSeek V4 Pro 0813 |
| Code migration (Vals AI) | 41.5% | 29.9% | DeepSeek V4 Pro 0813 |
| ProgramBench (Vals AI) | 0.0% | 0.5% | Muse Spark 1.2 |
| IOI (Vals AI) | 51.6% | 21.8% | DeepSeek V4 Pro 0813 |
| Terminal-Bench 2.1 (Vals AI) | 54.7% | 69.7% | Muse Spark 1.2 |
| Vibe Code Bench 1-100 (Vals AI) | 17.5% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| SkillsBench (Vals AI) | 53.8% | 53.0% | DeepSeek V4 Pro 0813 |
| CyberBench (Vals AI) | 72.9% | 69.5% | DeepSeek V4 Pro 0813 |
| SciCode | 51.0% | 57.4% | Muse Spark 1.2 |
| FrontierCode | 28.5% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| FrontierSWE | Not measured | 12.0% | Only Muse Spark 1.2 measured |
| WeirdML | 66.2% | 60.3% | DeepSeek V4 Pro 0813 |
| ALE-Bench | 1403.2 | Not measured | Only DeepSeek V4 Pro 0813 measured |
| LMArena Text: coding | 1469 | 1505 | Muse Spark 1.2 |
| LMArena Text: software and IT | 1462 | 1505 | Muse Spark 1.2 |
| LMArena WebDev: React | 1588 | 1526 | DeepSeek V4 Pro 0813 |
| LMArena WebDev: HTML | 1566 | 1543 | DeepSeek V4 Pro 0813 |
| Neither model has a result on the other 15 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index11
| Benchmark | DeepSeek V4 Pro 0813 | Muse Spark 1.2 | Leader |
|---|---|---|---|
| AA Intelligence Index | 36.0 | 39.6 | Muse Spark 1.2 |
| Epoch Capabilities Index | 155.6 | 155.1 | DeepSeek V4 Pro 0813 |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index25
| Benchmark | DeepSeek V4 Pro 0813 | Muse Spark 1.2 | Leader |
|---|---|---|---|
| AA Agentic Index | 41.3 | 43.2 | Muse Spark 1.2 |
| Design Arena: agents building web apps | Not measured | 1222 | Only Muse Spark 1.2 measured |
| Design Arena: agents building full-stack apps | Not measured | 1210 | Only Muse Spark 1.2 measured |
| Design Arena: agents building mobile apps | Not measured | 1182 | Only Muse Spark 1.2 measured |
| Design Arena: agents building Android apps | Not measured | 1228 | Only Muse Spark 1.2 measured |
| Design Arena: agents building games | Not measured | 1179 | Only Muse Spark 1.2 measured |
| Design Arena: HTML slides | Not measured | 1135 | Only Muse Spark 1.2 measured |
| Design Arena: PowerPoint slides | Not measured | 1198 | Only Muse Spark 1.2 measured |
| LiveBench research | Not measured | 67.7% | Only Muse Spark 1.2 measured |
| LiveBench browser | Not measured | 36.0% | Only Muse Spark 1.2 measured |
| LiveBench long agentic | Not measured | 21.3% | Only Muse Spark 1.2 measured |
| SREBench (Vals AI) | 1.9% | 0.0% | DeepSeek V4 Pro 0813 |
| Finance Agent (Vals AI) | 50.4% | 60.6% | Muse Spark 1.2 |
| Tax Agent Bench (Vals AI) | 26.8% | 32.8% | Muse Spark 1.2 |
| Legal Research Bench (Vals AI) | 40.9% | 43.8% | Muse Spark 1.2 |
| Harvey's Legal Agent Benchmark (Vals AI) | 10.8% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| Excel Modeling Benchmark (Vals AI) | 52.8% | 57.0% | Muse Spark 1.2 |
| BioMysteryBench (Vals AI) | Not measured | 64.8% | Only Muse Spark 1.2 measured |
| Terminal-Bench-Science (Vals AI) | 4.3% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| APEX-Agents | 47.3% | 45.4% | DeepSeek V4 Pro 0813 |
| Neither model has a result on the other 19 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index24
| Benchmark | DeepSeek V4 Pro 0813 | Muse Spark 1.2 | Leader |
|---|---|---|---|
| GPQA Diamond | 92.4% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| LiveBench reasoning | Not measured | 90.0% | Only Muse Spark 1.2 measured |
| MMMU Pro (Vals AI) | Not measured | 86.1% | Only Muse Spark 1.2 measured |
| GPQA Diamond (Artificial Analysis run) | 92.8% | 90.4% | DeepSeek V4 Pro 0813 |
| Humanity's Last Exam (Artificial Analysis run) | 41.0% | 45.5% | Muse Spark 1.2 |
| GPQA Diamond (Epoch AI run) | 91.7% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| ARC-AGI | 90.5% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| ARC-AGI-2 | 61.3% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| SimpleBench | Not measured | 74.5% | Only Muse Spark 1.2 measured |
| Chess puzzles | 46.0% | 27.0% | DeepSeek V4 Pro 0813 |
| Mystery game puzzles | 43.0% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| LMArena Text: hard prompts | 1458 | 1489 | Muse Spark 1.2 |
| LMArena Text: expert | 1459 | 1475 | Muse Spark 1.2 |
| LMArena Text: hard prompts in English | 1463 | 1487 | Muse Spark 1.2 |
| LMArena Vision: diagrams | Not measured | 1317 | Only Muse Spark 1.2 measured |
| Neither model has a result on the other 6 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index12
| Benchmark | DeepSeek V4 Pro 0813 | Muse Spark 1.2 | Leader |
|---|---|---|---|
| OTIS Mock AIME | 98.6% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| FrontierMath, tiers 1 to 3 | 64.6% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| FrontierMath, tier 4 | 26.8% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| ProofBench | 50.0% | 43.0% | DeepSeek V4 Pro 0813 |
| LMArena Text: math | 1441 | 1484 | Muse Spark 1.2 |
| LMArena Text: mathematical work | 1443 | 1480 | Muse Spark 1.2 |
| Neither model has a result on the other 7 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index010
| Benchmark | DeepSeek V4 Pro 0813 | Muse Spark 1.2 | Leader |
|---|---|---|---|
| MMLU Pro | 87.0% | 88.3% | Muse Spark 1.2 |
| LegalBench (Vals AI) | 82.4% | 85.3% | Muse Spark 1.2 |
| CorpFin (Vals AI) | 65.4% | 70.9% | Muse Spark 1.2 |
| TaxEval (Vals AI) | 73.1% | 80.4% | Muse Spark 1.2 |
| MortgageTax (Vals AI) | Not measured | 65.4% | Only Muse Spark 1.2 measured |
| MedCode (Vals AI) | 42.5% | 49.4% | Muse Spark 1.2 |
| SimpleQA Verified | 52.9% | 60.3% | Muse Spark 1.2 |
| LMArena Text: science | 1457 | 1500 | Muse Spark 1.2 |
| LMArena Text: medicine and healthcare | 1446 | 1496 | Muse Spark 1.2 |
| LMArena Text: legal and government | 1459 | 1517 | Muse Spark 1.2 |
| LMArena Text: business and finance | 1442 | 1484 | Muse Spark 1.2 |
| Neither model has a result on the other 6 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index9 boards, none shared
| Benchmark | DeepSeek V4 Pro 0813 | Muse Spark 1.2 | Leader |
|---|---|---|---|
| Design Arena: websites | Not measured | 1313 | Only Muse Spark 1.2 measured |
| Design Arena: UI components | Not measured | 1300 | Only Muse Spark 1.2 measured |
| Design Arena: game dev | Not measured | 1302 | Only Muse Spark 1.2 measured |
| Design Arena: data viz | Not measured | 1334 | Only Muse Spark 1.2 measured |
| Design Arena: 3D | Not measured | 1301 | Only Muse Spark 1.2 measured |
| Design Arena: all code categories | Not measured | 1313 | Only Muse Spark 1.2 measured |
| Design Arena: SVG | Not measured | 1285 | Only Muse Spark 1.2 measured |
| Design Arena: ASCII art | Not measured | 1284 | Only Muse Spark 1.2 measured |
| LiveBench design | Not measured | 67.1% | Only Muse Spark 1.2 measured |
GeneralLike-for-likePublic boards, shown as published, kept out of the index216
| Benchmark | DeepSeek V4 Pro 0813 | Muse Spark 1.2 | Leader |
|---|---|---|---|
| Vals Index | 53.0% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| LiveBench data analysis | Not measured | 76.5% | Only Muse Spark 1.2 measured |
| Public Benefits Bench (Vals AI) | Not measured | 68.5% | Only Muse Spark 1.2 measured |
| SAGE (Vals AI) | Not measured | 47.7% | Only Muse Spark 1.2 measured |
| MedScribe (Vals AI) | 80.2% | 90.1% | Muse Spark 1.2 |
| Vals Multimodal Index | Not measured | 69.8% | Only Muse Spark 1.2 measured |
| Long context reasoning (AA-LCR) | 80.3% | 79.0% | DeepSeek V4 Pro 0813 |
| DTBench | 93.9% | 94.7% | Muse Spark 1.2 |
| LMCA | 45.5% | 48.4% | Muse Spark 1.2 |
| LMArena Text | 1446 | 1482 | Muse Spark 1.2 |
| LMArena Text: instruction following | 1448 | 1462 | Muse Spark 1.2 |
| LMArena Text: creative writing | 1439 | 1436 | DeepSeek V4 Pro 0813 |
| LMArena Vision | Not measured | 1304 | Only Muse Spark 1.2 measured |
| LMArena Text: English | 1457 | 1483 | Muse Spark 1.2 |
| LMArena Text: non-English | 1431 | 1476 | Muse Spark 1.2 |
| LMArena Text: Chinese | 1481 | 1509 | Muse Spark 1.2 |
| LMArena Text: Korean | 1449 | 1462 | Muse Spark 1.2 |
| LMArena Text: French | 1446 | 1495 | Muse Spark 1.2 |
| LMArena Text: Spanish | 1435 | 1484 | Muse Spark 1.2 |
| LMArena Text: Russian | 1448 | 1483 | Muse Spark 1.2 |
| LMArena Text: multi-turn | 1454 | 1502 | Muse Spark 1.2 |
| LMArena Text: longer queries | 1458 | 1473 | Muse Spark 1.2 |
| LMArena Text: writing and language | 1443 | 1453 | Muse Spark 1.2 |
| LMArena Text: entertainment and media | 1424 | 1443 | Muse Spark 1.2 |
| LMArena Vision: English | Not measured | 1291 | Only Muse Spark 1.2 measured |
| LMArena Vision: reading text | Not measured | 1313 | Only Muse Spark 1.2 measured |
| Neither model has a result on the other 20 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | DeepSeek V4 Pro 0813 | Muse Spark 1.2 | Leader |
|---|---|---|---|
| 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 0813 costs less
Repository review
50K input + 3K output tokens
DeepSeek V4 Pro 0813 costs less
Agent loop
200K input + 10K output tokens
DeepSeek V4 Pro 0813 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.
| Specification | DeepSeek V4 Pro 0813 | Meta Muse Spark 1.2 |
|---|---|---|
| Maker | DeepSeek | Meta |
| Weights | Open weights | No public weights listed |
| Released | August 12, 2026 | August 5, 2026 |
| Context window | 1.05M | 1.05M |
| Input $/Mtok | $0.66 | $1.25 |
| Output $/Mtok | $1.98 | $4.25 |
| Coding index | 54.6 | 65.1 |
| Independent measurements behind it | 1 | 2 |
| Cost per solved task | $0.10 | $0.64 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is DeepSeek V4 Pro 0813 better than Muse Spark 1.2 for coding?
Pick Muse Spark 1.2 for coding: it scores 65.1 on the Vibe Coding Index against 54.6 for DeepSeek V4 Pro 0813. Pick DeepSeek V4 Pro 0813 when output tokens cost 2.1x less, or when a solved task costs $0.10 against $0.64.
Which is cheaper, DeepSeek V4 Pro 0813 or Muse Spark 1.2?
DeepSeek V4 Pro 0813 costs $0.66 per million input tokens and $1.98 per million output tokens; Muse Spark 1.2 costs $1.25 and $4.25. Per solved SWE-bench Verified task, DeepSeek V4 Pro 0813 costs $0.10 and Muse Spark 1.2 $0.64.
Which has the bigger context window, DeepSeek V4 Pro 0813 or Muse Spark 1.2?
Neither: both take 1.05M tokens.
How many benchmarks is the DeepSeek V4 Pro 0813 vs Muse Spark 1.2 comparison based on?
63 benchmarks have a result for both models as of October 21, 2026. DeepSeek V4 Pro 0813's coding index rests on 1 independent benchmark and Muse Spark 1.2's on 2. Where only one model has a result, the page says so instead of naming a winner.