# DeepSeek LLM 67B vs Hy4 preview

> Hy4 preview is the stronger model overall, scoring 45.3 to 24.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-llm-67b-vs-hy4-preview
- Last updated: 2026-10-10
- Shared benchmarks: 0

## Summary

- The widest gap is in math, where Hy4 preview leads 55.7 to 8.7.

## Snapshot

| | DeepSeek LLM 67B | Hy4 preview |
|---|---|---|
| Provider | DeepSeek | Tencent |
| Noometry Index | 24.9 | 45.3 |
| Rank | 347 | 73 |
| Context | — | 1.05M |
| Input $/M | — | $0.75 |
| Output $/M | — | $2.25 |
| Weights | Open | Open |

## Coding

- DeepSeek LLM 67B: 31.9 (#278)
- Hy4 preview: 51.6 (#38)

| Benchmark | DeepSeek LLM 67B | Hy4 preview |
|---|---|---|
| LMArena WebDev | — | 1632 |
| LMArena Coding | 1096 | — |

## Reasoning

- DeepSeek LLM 67B: 16.5 (#304)
- Hy4 preview: 31.9 (#79)

| Benchmark | DeepSeek LLM 67B | Hy4 preview |
|---|---|---|
| NYT Connections (extended) | — | 68.2% |
| Chess Puzzles | 0% | — |
| LMArena Hard Prompts | 1070 | — |
| Epoch Capabilities Index | 110.5 | — |

## Math

- DeepSeek LLM 67B: 8.7 (#324)
- Hy4 preview: 55.7 (#42)

| Benchmark | DeepSeek LLM 67B | Hy4 preview |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.8% | — |
| ProofBench | — | 75% |
| LMArena Math | 1108 | — |
| MATH Level 5 | 6.4% | — |

## Knowledge

- DeepSeek LLM 67B: 7.0 (#313)
- Hy4 preview: —

| Benchmark | DeepSeek LLM 67B | Hy4 preview |
|---|---|---|
| GPQA Diamond | 24.6% | — |

## Multilingual

- DeepSeek LLM 67B: 29.4 (#267)
- Hy4 preview: —

| Benchmark | DeepSeek LLM 67B | Hy4 preview |
|---|---|---|
| LMArena Non-English | 1073 | — |
| LMArena Chinese | 1132 | — |

## Instruction Following

- DeepSeek LLM 67B: 55.4 (#277)
- Hy4 preview: —

| Benchmark | DeepSeek LLM 67B | Hy4 preview |
|---|---|---|
| LMArena Instruction Following | 1079 | — |

## Long Context

- DeepSeek LLM 67B: 33.1 (#265)
- Hy4 preview: —

| Benchmark | DeepSeek LLM 67B | Hy4 preview |
|---|---|---|
| LMArena Longer Query | 1092 | — |

## Writing & Preference

- DeepSeek LLM 67B: 31.6 (#282)
- Hy4 preview: —

| Benchmark | DeepSeek LLM 67B | Hy4 preview |
|---|---|---|
| LMArena Text | 1105 | — |
| LMArena Creative Writing | 1067 | — |
| LMArena Multi-Turn | 1082 | — |

## FAQ

### Is DeepSeek LLM 67B better than Hy4 preview?

Hy4 preview is the stronger model overall, scoring 45.3 to 24.9 on the Noometry Index.

### Is DeepSeek LLM 67B or Hy4 preview better for coding?

Hy4 preview scores higher on coding benchmarks: 51.6 versus 31.9 in the Noometry coding category.

### How many benchmarks do DeepSeek LLM 67B and Hy4 preview share?

0 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and Hy4 preview has 3.