# Hy4 preview vs Muse Spark 1.2

> Muse Spark 1.2 is the stronger model overall, scoring 49.8 to 44.6 on the Noometry Index. Hy4 preview costs 1.6× less per token, which makes it the better buy when Muse Spark 1.2's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/hy4-preview-vs-muse-spark-1-2
- Last updated: 2026-10-10
- Shared benchmarks: 3

## Summary

- They share 3 benchmarks with published results for both. Hy4 preview scores higher in 2 categories and Muse Spark 1.2 in 1 category; 3 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Muse Spark 1.2 leads 50.4 to 30.6.
- The biggest single-benchmark swing is ProofBench: 75% for Hy4 preview and 43% for Muse Spark 1.2.
- Hy4 preview is cheaper at $0.83 / $2.50 per million input/output tokens, against $1.25 / $4.25 for Muse Spark 1.2.
- Hy4 preview has downloadable open weights; the other is API-only.

## Snapshot

| | Hy4 preview | Muse Spark 1.2 |
|---|---|---|
| Provider | Tencent | Meta |
| Noometry Index | 44.6 | 49.8 |
| Rank | 73 | 44 |
| Context | 1.05M | 1.05M |
| Input $/M | $0.83 | $1.25 |
| Output $/M | $2.50 | $4.25 |
| Weights | Open | Proprietary |

## Coding

- Hy4 preview: 51.4 (#36)
- Muse Spark 1.2: 48.6 (#49)

| Benchmark | Hy4 preview | Muse Spark 1.2 |
|---|---|---|
| LMArena WebDev | 1633 | 1532 |
| DeepSWE | — | 54.9% |
| FrontierSWE | — | 12% |
| SciCode | — | 56.4% |
| WeirdML | — | 60.3% |
| LMArena Coding | — | 1497 |

## Agentic & Tool Use

- Hy4 preview: —
- Muse Spark 1.2: 29.5 (#77)

| Benchmark | Hy4 preview | Muse Spark 1.2 |
|---|---|---|
| APEX-Agents | — | 36.4% |
| GDP.pdf | — | 16% |

## Reasoning

- Hy4 preview: 30.6 (#75)
- Muse Spark 1.2: 50.4 (#32)

| Benchmark | Hy4 preview | Muse Spark 1.2 |
|---|---|---|
| NYT Connections (extended) | 68.2% | 79.2% |
| SimpleBench | — | 74.5% |
| CritPt | — | 17.7% |
| LMArena Hard Prompts | — | 1490 |
| DTBench | — | 94.7% |
| LMCA | — | 48.4% |
| Epoch Capabilities Index | — | 155.18 |

## Math

- Hy4 preview: 54.5 (#37)
- Muse Spark 1.2: 43.8 (#68)

| Benchmark | Hy4 preview | Muse Spark 1.2 |
|---|---|---|
| ProofBench | 75% | 43% |

## Knowledge

- Hy4 preview: —
- Muse Spark 1.2: 53.9 (#49)

| Benchmark | Hy4 preview | Muse Spark 1.2 |
|---|---|---|
| SimpleQA Verified | — | 60.3% |
| LMArena Expert | — | 1485 |

## Multimodal

- Hy4 preview: —
- Muse Spark 1.2: 42.6 (#24)

| Benchmark | Hy4 preview | Muse Spark 1.2 |
|---|---|---|
| LMArena Vision | — | 1304 |

## Multilingual

- Hy4 preview: —
- Muse Spark 1.2: 57.4 (#10)

| Benchmark | Hy4 preview | Muse Spark 1.2 |
|---|---|---|
| LMArena Non-English | — | 1479 |
| LMArena Chinese | — | 1519 |
| LMArena Russian | — | 1494 |

## Instruction Following

- Hy4 preview: —
- Muse Spark 1.2: 77.1 (#30)

| Benchmark | Hy4 preview | Muse Spark 1.2 |
|---|---|---|
| LMArena Instruction Following | — | 1466 |

## Long Context

- Hy4 preview: —
- Muse Spark 1.2: 45.4 (#35)

| Benchmark | Hy4 preview | Muse Spark 1.2 |
|---|---|---|
| LMArena Longer Query | — | 1483 |

## Writing & Preference

- Hy4 preview: —
- Muse Spark 1.2: 72.6 (#14)

| Benchmark | Hy4 preview | Muse Spark 1.2 |
|---|---|---|
| LMArena Text | — | 1486 |
| LMArena Creative Writing | — | 1444 |
| EQ-Bench Creative Writing | — | 1840 |
| LMArena Multi-Turn | — | 1504 |

## FAQ

### Is Hy4 preview better than Muse Spark 1.2?

Muse Spark 1.2 is the stronger model overall, scoring 49.8 to 44.6 on the Noometry Index. Hy4 preview costs 1.6× less per token, which makes it the better buy when Muse Spark 1.2's lead doesn't matter for your workload.

### Which is cheaper, Hy4 preview or Muse Spark 1.2?

Hy4 preview is cheaper. It lists at $0.83 per million input tokens and $2.50 per million output tokens; Muse Spark 1.2 lists at $1.25 and $4.25.

### Is Hy4 preview or Muse Spark 1.2 better for coding?

Hy4 preview scores higher on coding benchmarks: 51.4 versus 48.6 in the Noometry coding category.

### Which has the bigger context window?

Both accept 1.05M tokens.

### How many benchmarks do Hy4 preview and Muse Spark 1.2 share?

3 benchmarks have published results for both models. Hy4 preview has 3 scored results on Noometry and Muse Spark 1.2 has 28.