# DeepSeek V4 Pro vs Muse Spark 1.2

> DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 50.3 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-v4-pro-vs-muse-spark-1-2
- Last updated: 2026-10-10
- Shared benchmarks: 26

## Summary

- They share 26 benchmarks with published results for both. DeepSeek V4 Pro scores higher in 5 categories and Muse Spark 1.2 in 4 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek V4 Pro leads 64.8 to 46.4.
- The biggest single-benchmark swing is NYT Connections (extended): 91.3% for DeepSeek V4 Pro and 79.2% for Muse Spark 1.2.
- DeepSeek V4 Pro is cheaper at $0.66 / $1.98 per million input/output tokens, against $1.25 / $4.25 for Muse Spark 1.2.
- Muse Spark 1.2 accepts more context: 1.05M tokens versus 1M.
- DeepSeek V4 Pro has downloadable open weights; the other is API-only.

## Snapshot

| | DeepSeek V4 Pro | Muse Spark 1.2 |
|---|---|---|
| Provider | DeepSeek | Meta |
| Noometry Index | 54.3 | 50.3 |
| Rank | 31 | 48 |
| Context | 1M | 1.05M |
| Input $/M | $0.66 | $1.25 |
| Output $/M | $1.98 | $4.25 |
| Weights | Open | Proprietary |

## Coding

- DeepSeek V4 Pro: 52.4 (#34)
- Muse Spark 1.2: 49.2 (#51)

| Benchmark | DeepSeek V4 Pro | Muse Spark 1.2 |
|---|---|---|
| LMArena WebDev | 1582 | 1533 |
| SciCode | 51% | 56.4% |
| WeirdML | 66.2% | 60.3% |
| LMArena Coding | 1470 | 1495 |
| SWE-bench Verified | 77.6% | — |
| DeepSWE | — | 54.9% |
| FrontierCode | 28.6% | — |
| FrontierSWE | — | 12% |
| ALE-Bench | 1,403 | — |

## Agentic & Tool Use

- DeepSeek V4 Pro: 32.8 (#58)
- Muse Spark 1.2: 29.4 (#87)

| Benchmark | DeepSeek V4 Pro | Muse Spark 1.2 |
|---|---|---|
| APEX-Agents | 47.3% | 36.4% |
| GDP.pdf | — | 16% |
| Vending-Bench 2 | 3,285 | — |

## Reasoning

- DeepSeek V4 Pro: 56.5 (#24)
- Muse Spark 1.2: 51.3 (#34)

| Benchmark | DeepSeek V4 Pro | Muse Spark 1.2 |
|---|---|---|
| NYT Connections (extended) | 91.3% | 79.2% |
| CritPt | 18% | 17.7% |
| LMArena Hard Prompts | 1461 | 1486 |
| DTBench | 93.9% | 94.7% |
| LMCA | 45.5% | 48.4% |
| Epoch Capabilities Index | 155.31 | 154.87 |
| ARC-AGI-2 | 61.3% | — |
| SimpleBench | — | 74.5% |
| Kagi LLM Benchmark | 53.5% | — |
| ARC-AGI-1 | 90.5% | — |
| Chess Puzzles | 47% | — |
| Mystery Game Puzzles | 43% | — |
| Surface Evolver Bench | 40% | — |
| ForecastBench | 56.1 | — |

## Math

- DeepSeek V4 Pro: 64.8 (#30)
- Muse Spark 1.2: 46.4 (#70)

| Benchmark | DeepSeek V4 Pro | Muse Spark 1.2 |
|---|---|---|
| ProofBench | 50% | 43% |
| LMArena Math | 1455 | 1471 |
| FrontierMath (Tiers 1-3) | 64.6% | — |
| FrontierMath Tier 4 | 26.8% | — |
| MathArena Final-Answer Competitions | 76.6% | — |
| OTIS Mock AIME 2024-2025 | 98.6% | — |

## Knowledge

- DeepSeek V4 Pro: 59.5 (#31)
- Muse Spark 1.2: 54.1 (#53)

| Benchmark | DeepSeek V4 Pro | Muse Spark 1.2 |
|---|---|---|
| SimpleQA Verified | 52.9% | 60.3% |
| LMArena Expert | 1464 | 1480 |
| GPQA Diamond | 91.7% | — |
| Vectara Hallucination Rate | 8.6% | — |

## Multimodal

- DeepSeek V4 Pro: —
- Muse Spark 1.2: 43.4 (#25)

| Benchmark | DeepSeek V4 Pro | Muse Spark 1.2 |
|---|---|---|
| LMArena Vision | — | 1305 |

## Multilingual

- DeepSeek V4 Pro: 54.4 (#45)
- Muse Spark 1.2: 57.1 (#11)

| Benchmark | DeepSeek V4 Pro | Muse Spark 1.2 |
|---|---|---|
| LMArena Non-English | 1439 | 1478 |
| LMArena Chinese | 1486 | 1511 |
| LMArena French | 1472 | 1513 |
| LMArena Russian | 1453 | 1487 |
| LMArena Spanish | 1458 | 1498 |
| LMArena German | 1458 | — |
| LMArena Japanese | 1445 | — |
| LMArena Korean | 1447 | — |

## Instruction Following

- DeepSeek V4 Pro: 76.1 (#47)
- Muse Spark 1.2: 76.7 (#36)

| Benchmark | DeepSeek V4 Pro | Muse Spark 1.2 |
|---|---|---|
| LMArena Instruction Following | 1448 | 1461 |

## Long Context

- DeepSeek V4 Pro: 45.0 (#51)
- Muse Spark 1.2: 45.2 (#48)

| Benchmark | DeepSeek V4 Pro | Muse Spark 1.2 |
|---|---|---|
| LMArena Longer Query | 1458 | 1475 |
| CL-bench Life | 13.5% | — |

## Writing & Preference

- DeepSeek V4 Pro: 65.5 (#46)
- Muse Spark 1.2: 72.3 (#14)

| Benchmark | DeepSeek V4 Pro | Muse Spark 1.2 |
|---|---|---|
| LMArena Text | 1451 | 1482 |
| LMArena Creative Writing | 1446 | 1449 |
| EQ-Bench Creative Writing | 1553 | 1840 |
| LMArena Multi-Turn | 1467 | 1494 |
| EQ-Bench 4 | 1166 | — |

## FAQ

### Is DeepSeek V4 Pro better than Muse Spark 1.2?

DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 50.3 on the Noometry Index.

### Which is cheaper, DeepSeek V4 Pro or Muse Spark 1.2?

DeepSeek V4 Pro is cheaper. It lists at $0.66 per million input tokens and $1.98 per million output tokens; Muse Spark 1.2 lists at $1.25 and $4.25.

### Is DeepSeek V4 Pro or Muse Spark 1.2 better for coding?

DeepSeek V4 Pro scores higher on coding benchmarks: 52.4 versus 49.2 in the Noometry coding category.

### Which has the bigger context window?

Muse Spark 1.2 does, with 1.05M tokens against 1M.

### How many benchmarks do DeepSeek V4 Pro and Muse Spark 1.2 share?

26 benchmarks have published results for both models. DeepSeek V4 Pro has 48 scored results on Noometry and Muse Spark 1.2 has 31.