# Hy3 vs Mixtral 8x7B

> Hy3 is the stronger model overall, scoring 44.2 to 27.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/hy3-vs-mixtral-8x7b
- Last updated: 2026-10-10
- Shared benchmarks: 17

## Summary

- They share 17 benchmarks with published results for both. Hy3 scores higher in 8 categories and Mixtral 8x7B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Hy3 leads 40.8 to 11.0.
- Hy3 is cheaper at $0.0825 / $0.33 per million input/output tokens, against $0.70 / $0.70 for Mixtral 8x7B.
- Hy3 accepts more context: 262K tokens versus 32K.

## Snapshot

| | Hy3 | Mixtral 8x7B |
|---|---|---|
| Provider | Tencent | Mistral AI |
| Noometry Index | 44.2 | 27.1 |
| Rank | 79 | 334 |
| Context | 262K | 32K |
| Input $/M | $0.0825 | $0.70 |
| Output $/M | $0.33 | $0.70 |
| Weights | Open | Open |

## Coding

- Hy3: 46.8 (#63)
- Mixtral 8x7B: 32.8 (#269)

| Benchmark | Hy3 | Mixtral 8x7B |
|---|---|---|
| LMArena Coding | 1464 | 1126 |
| LMArena WebDev | 1508 | — |
| HumanEval+ | — | 39.6% |
| MBPP+ | — | 49.7% |

## Reasoning

- Hy3: 26.1 (#136)
- Mixtral 8x7B: 18.2 (#285)

| Benchmark | Hy3 | Mixtral 8x7B |
|---|---|---|
| LMArena Hard Prompts | 1447 | 1115 |
| NYT Connections (extended) | 41.2% | — |
| DTBench | — | 49.6% |
| Adversarial NLI | — | 55.2% |
| Epoch Capabilities Index | — | 118.47 |
| ForecastBench | — | 56.3 |
| HellaSwag | — | 86.7% |
| PIQA | — | 83.6% |
| WinoGrande | — | 77.2% |

## Math

- Hy3: 40.1 (#93)
- Mixtral 8x7B: 18.8 (#289)

| Benchmark | Hy3 | Mixtral 8x7B |
|---|---|---|
| LMArena Math | 1475 | 1147 |
| Omni-MATH | — | 10.5% |
| MATH Level 5 | — | 10% |
| GSM8K | — | 74.4% |

## Knowledge

- Hy3: 40.8 (#114)
- Mixtral 8x7B: 11.0 (#301)

| Benchmark | Hy3 | Mixtral 8x7B |
|---|---|---|
| LMArena Expert | 1460 | 1088 |
| GPQA Diamond | — | 30.6% |
| MMLU-Pro | — | 33.5% |
| GPQA (HELM) | — | 29.6% |
| ARC (AI2) Challenge | — | 87.3% |
| MMLU | — | 70.6% |
| OpenBookQA | — | 85.8% |
| TriviaQA | — | 82.2% |

## Multilingual

- Hy3: 53.5 (#65)
- Mixtral 8x7B: 29.6 (#266)

| Benchmark | Hy3 | Mixtral 8x7B |
|---|---|---|
| LMArena Non-English | 1426 | 1077 |
| LMArena Chinese | 1493 | 1055 |
| LMArena French | 1461 | 1166 |
| LMArena German | 1439 | 1114 |
| LMArena Japanese | 1392 | 931 |
| LMArena Korean | 1395 | 968 |
| LMArena Russian | 1432 | 1090 |
| LMArena Spanish | 1456 | 1111 |

## Instruction Following

- Hy3: 75.1 (#70)
- Mixtral 8x7B: 51.0 (#297)

| Benchmark | Hy3 | Mixtral 8x7B |
|---|---|---|
| LMArena Instruction Following | 1426 | 1109 |
| IFEval | — | 57.5% |

## Long Context

- Hy3: 44.1 (#75)
- Mixtral 8x7B: 33.4 (#260)

| Benchmark | Hy3 | Mixtral 8x7B |
|---|---|---|
| LMArena Longer Query | 1442 | 1103 |

## Writing & Preference

- Hy3: 62.2 (#81)
- Mixtral 8x7B: 34.2 (#270)

| Benchmark | Hy3 | Mixtral 8x7B |
|---|---|---|
| LMArena Text | 1439 | 1132 |
| LMArena Creative Writing | 1402 | 1109 |
| LMArena Multi-Turn | 1436 | 1115 |
| WildBench | — | 67.3% |

## FAQ

### Is Hy3 better than Mixtral 8x7B?

Hy3 is the stronger model overall, scoring 44.2 to 27.1 on the Noometry Index.

### Which is cheaper, Hy3 or Mixtral 8x7B?

Hy3 is cheaper. It lists at $0.0825 per million input tokens and $0.33 per million output tokens; Mixtral 8x7B lists at $0.70 and $0.70.

### Is Hy3 or Mixtral 8x7B better for coding?

Hy3 scores higher on coding benchmarks: 46.8 versus 32.8 in the Noometry coding category.

### Which has the bigger context window?

Hy3 does, with 262K tokens against 32K.

### How many benchmarks do Hy3 and Mixtral 8x7B share?

17 benchmarks have published results for both models. Hy3 has 19 scored results on Noometry and Mixtral 8x7B has 38.