# Muse Spark 1.3 vs Phi-4

> Muse Spark 1.3 is the stronger model overall, scoring 54.8 to 31.2 on the Noometry Index. Phi-4 costs 23× less per token, which makes it the better buy when Muse Spark 1.3's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/muse-spark-1-3-vs-phi-4
- Last updated: 2026-10-10
- Shared benchmarks: 20

## Summary

- They share 20 benchmarks with published results for both. Muse Spark 1.3 scores higher in 9 categories and Phi-4 in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Muse Spark 1.3 leads 73.1 to 20.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 99.2% for Muse Spark 1.3 and 13.8% for Phi-4.
- Phi-4 is cheaper at $0.07 / $0.14 per million input/output tokens, against $1.25 / $4.25 for Muse Spark 1.3.
- Muse Spark 1.3 accepts more context: 1.05M tokens versus 128K.
- Phi-4 has downloadable open weights; the other is API-only.

## Snapshot

| | Muse Spark 1.3 | Phi-4 |
|---|---|---|
| Provider | Meta | Microsoft |
| Noometry Index | 54.8 | 31.2 |
| Rank | 27 | 279 |
| Context | 1.05M | 128K |
| Input $/M | $1.25 | $0.07 |
| Output $/M | $4.25 | $0.14 |
| Weights | Proprietary | Open |

## Coding

- Muse Spark 1.3: 56.6 (#21)
- Phi-4: 34.4 (#239)

| Benchmark | Muse Spark 1.3 | Phi-4 |
|---|---|---|
| LMArena Coding | 1514 | 1231 |
| CursorBench | 41.6% | — |
| LMArena WebDev | 1657 | — |
| SciCode | 59.7% | — |
| BigCodeBench Instruct | — | 45.5% |
| LiveBench Coding | — | 30.7% |
| BigCodeBench Complete | — | 55.4% |

## Agentic & Tool Use

- Muse Spark 1.3: 38.6 (#30)
- Phi-4: 22.8 (#128)

| Benchmark | Muse Spark 1.3 | Phi-4 |
|---|---|---|
| APEX-Agents | 57.8% | — |
| Berkeley Function Calling Leaderboard | — | 28.8% |
| BALROG | — | 11.6% |
| GDP.pdf | 27.6% | — |

## Reasoning

- Muse Spark 1.3: 54.0 (#27)
- Phi-4: 17.7 (#291)

| Benchmark | Muse Spark 1.3 | Phi-4 |
|---|---|---|
| Chess Puzzles | 38% | 1% |
| LMArena Hard Prompts | 1503 | 1220 |
| Epoch Capabilities Index | 156.75 | 130.42 |
| NYT Connections (extended) | 85.1% | — |
| CritPt | 26% | — |
| LiveBench Reasoning | — | 47.8% |
| Mystery Game Puzzles | 25% | — |
| DTBench | 96.5% | — |
| LiveBench Data Analysis | — | 45.2% |
| LMCA | 53.9% | — |
| Bench to the Future 3 | 0.14 | — |
| LiveBench | — | 41.6% |

## Math

- Muse Spark 1.3: 73.1 (#21)
- Phi-4: 20.8 (#285)

| Benchmark | Muse Spark 1.3 | Phi-4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 99.2% | 13.8% |
| LMArena Math | 1494 | 1246 |
| FrontierMath (Tiers 1-3) | 74.4% | — |
| FrontierMath Tier 4 | 46.3% | — |
| ProofBench | 58% | — |
| LiveBench Math | — | 42% |
| MATH Level 5 | — | 64.9% |

## Knowledge

- Muse Spark 1.3: 42.6 (#95)
- Phi-4: 32.6 (#209)

| Benchmark | Muse Spark 1.3 | Phi-4 |
|---|---|---|
| LMArena Expert | 1516 | 1203 |
| GPQA Diamond | — | 56.1% |
| Confabulations | — | 29.4% |
| Vectara Hallucination Rate | — | 3.7% |
| MMLU | — | 84.8% |

## Multimodal

- Muse Spark 1.3: 43.7 (#22)
- Phi-4: —

| Benchmark | Muse Spark 1.3 | Phi-4 |
|---|---|---|
| LMArena Vision | 1309 | — |
| LMArena Document | 1471 | — |

## Multilingual

- Muse Spark 1.3: 57.4 (#8)
- Phi-4: 37.2 (#237)

| Benchmark | Muse Spark 1.3 | Phi-4 |
|---|---|---|
| LMArena Non-English | 1481 | 1197 |
| LMArena Chinese | 1529 | 1212 |
| LMArena French | 1524 | 1224 |
| LMArena German | 1515 | 1222 |
| LMArena Japanese | 1474 | 1158 |
| LMArena Korean | 1501 | 1151 |
| LMArena Russian | 1490 | 1209 |
| LMArena Spanish | 1490 | 1234 |

## Instruction Following

- Muse Spark 1.3: 77.5 (#22)
- Phi-4: 60.4 (#251)

| Benchmark | Muse Spark 1.3 | Phi-4 |
|---|---|---|
| LMArena Instruction Following | 1477 | 1201 |
| LiveBench Instruction Following | — | 58.4% |

## Long Context

- Muse Spark 1.3: 45.6 (#32)
- Phi-4: 36.9 (#226)

| Benchmark | Muse Spark 1.3 | Phi-4 |
|---|---|---|
| LMArena Longer Query | 1488 | 1217 |

## Writing & Preference

- Muse Spark 1.3: 73.6 (#9)
- Phi-4: 40.5 (#244)

| Benchmark | Muse Spark 1.3 | Phi-4 |
|---|---|---|
| LMArena Text | 1490 | 1217 |
| LMArena Creative Writing | 1455 | 1182 |
| LMArena Multi-Turn | 1482 | 1206 |
| Short-Story Creative Writing | — | 62.6% |
| EQ-Bench Creative Writing | 1906 | — |
| LiveBench Language | — | 25.6% |

## FAQ

### Is Muse Spark 1.3 better than Phi-4?

Muse Spark 1.3 is the stronger model overall, scoring 54.8 to 31.2 on the Noometry Index. Phi-4 costs 23× less per token, which makes it the better buy when Muse Spark 1.3's lead doesn't matter for your workload.

### Which is cheaper, Muse Spark 1.3 or Phi-4?

Phi-4 is cheaper. It lists at $0.07 per million input tokens and $0.14 per million output tokens; Muse Spark 1.3 lists at $1.25 and $4.25.

### Is Muse Spark 1.3 or Phi-4 better for coding?

Muse Spark 1.3 scores higher on coding benchmarks: 56.6 versus 34.4 in the Noometry coding category.

### Which has the bigger context window?

Muse Spark 1.3 does, with 1.05M tokens against 128K.

### How many benchmarks do Muse Spark 1.3 and Phi-4 share?

20 benchmarks have published results for both models. Muse Spark 1.3 has 37 scored results on Noometry and Phi-4 has 37.