# Hy4 preview vs Llama 3-8B

> Hy4 preview is the stronger model overall, scoring 45.3 to 25.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/hy4-preview-vs-llama-3-8b
- Last updated: 2026-10-11
- Shared benchmarks: 0

## Summary

- The widest gap is in math, where Hy4 preview leads 55.7 to 8.8.

## Snapshot

| | Hy4 preview | Llama 3-8B |
|---|---|---|
| Provider | Tencent | Meta |
| Noometry Index | 45.3 | 25.5 |
| Rank | 73 | 344 |
| Context | 1.05M | — |
| Input $/M | $0.83 | — |
| Output $/M | $2.50 | — |
| Weights | Open | Open |

## Coding

- Hy4 preview: 51.6 (#38)
- Llama 3-8B: 31.0 (#289)

| Benchmark | Hy4 preview | Llama 3-8B |
|---|---|---|
| LMArena WebDev | 1632 | — |
| BigCodeBench Instruct | — | 31.9% |
| LMArena Coding | — | 1152 |
| BigCodeBench Complete | — | 36.9% |
| HumanEval+ | — | 56.7% |
| MBPP+ | — | 54.8% |

## Reasoning

- Hy4 preview: 31.9 (#79)
- Llama 3-8B: 14.3 (#326)

| Benchmark | Hy4 preview | Llama 3-8B |
|---|---|---|
| NYT Connections (extended) | 68.2% | — |
| Chess Puzzles | — | 0% |
| LMArena Hard Prompts | — | 1133 |
| DTBench | — | 43.9% |
| Adversarial NLI | — | 57.3% |
| Epoch Capabilities Index | — | 116.45 |
| ForecastBench | — | 58.6 |
| WinoGrande | — | 75.7% |

## Math

- Hy4 preview: 55.7 (#42)
- Llama 3-8B: 8.8 (#323)

| Benchmark | Hy4 preview | Llama 3-8B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 1.9% |
| ProofBench | 75% | — |
| LMArena Math | — | 1151 |
| MATH Level 5 | — | 6.1% |

## Knowledge

- Hy4 preview: —
- Llama 3-8B: 7.8 (#308)

| Benchmark | Hy4 preview | Llama 3-8B |
|---|---|---|
| GPQA Diamond | — | 26.1% |
| LMArena Expert | — | 1113 |
| ARC (AI2) Challenge | — | 82.8% |
| MMLU | — | 68.8% |
| OpenBookQA | — | 82.6% |
| TriviaQA | — | 67.7% |

## Multilingual

- Hy4 preview: —
- Llama 3-8B: 30.8 (#261)

| Benchmark | Hy4 preview | Llama 3-8B |
|---|---|---|
| LMArena Non-English | — | 1098 |
| LMArena Chinese | — | 1076 |
| LMArena French | — | 1159 |
| LMArena German | — | 1104 |
| LMArena Japanese | — | 967 |
| LMArena Korean | — | 1004 |
| LMArena Russian | — | 1109 |
| LMArena Spanish | — | 1173 |

## Instruction Following

- Hy4 preview: —
- Llama 3-8B: 58.4 (#260)

| Benchmark | Hy4 preview | Llama 3-8B |
|---|---|---|
| LMArena Instruction Following | — | 1127 |

## Long Context

- Hy4 preview: —
- Llama 3-8B: 34.2 (#251)

| Benchmark | Hy4 preview | Llama 3-8B |
|---|---|---|
| LMArena Longer Query | — | 1128 |

## Writing & Preference

- Hy4 preview: —
- Llama 3-8B: 37.5 (#256)

| Benchmark | Hy4 preview | Llama 3-8B |
|---|---|---|
| LMArena Text | — | 1166 |
| LMArena Creative Writing | — | 1150 |
| LMArena Multi-Turn | — | 1152 |

## FAQ

### Is Hy4 preview better than Llama 3-8B?

Hy4 preview is the stronger model overall, scoring 45.3 to 25.5 on the Noometry Index.

### Is Hy4 preview or Llama 3-8B better for coding?

Hy4 preview scores higher on coding benchmarks: 51.6 versus 31.0 in the Noometry coding category.

### How many benchmarks do Hy4 preview and Llama 3-8B share?

0 benchmarks have published results for both models. Hy4 preview has 3 scored results on Noometry and Llama 3-8B has 34.