# Deepseek Coder v2 vs Hy3

> Hy3 is the stronger model overall, scoring 43.6 to 35.3 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-coder-v2-vs-hy3
- Last updated: 2026-10-10
- Shared benchmarks: 15

## Summary

- They share 15 benchmarks with published results for both. Deepseek Coder v2 scores higher in 0 categories and Hy3 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Hy3 leads 62.5 to 38.4.

## Snapshot

| | Deepseek Coder v2 | Hy3 |
|---|---|---|
| Provider | DeepSeek | Tencent |
| Noometry Index | 35.3 | 43.6 |
| Rank | 217 | 79 |
| Context | — | 262K |
| Input $/M | — | $0.13 |
| Output $/M | — | $0.53 |
| Weights | Open | Open |

## Coding

- Deepseek Coder v2: 37.5 (#178)
- Hy3: 46.6 (#60)

| Benchmark | Deepseek Coder v2 | Hy3 |
|---|---|---|
| LMArena Coding | 1251 | 1465 |
| LMArena WebDev | — | 1509 |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 59.7% | — |
| HumanEval+ | 82.3% | — |
| MBPP+ | 75.1% | — |

## Reasoning

- Deepseek Coder v2: 22.6 (#175)
- Hy3: 25.0 (#134)

| Benchmark | Deepseek Coder v2 | Hy3 |
|---|---|---|
| LMArena Hard Prompts | 1207 | 1449 |
| NYT Connections (extended) | — | 41.2% |
| WinoGrande | 83.7% | — |

## Math

- Deepseek Coder v2: 32.3 (#188)
- Hy3: 37.4 (#98)

| Benchmark | Deepseek Coder v2 | Hy3 |
|---|---|---|
| LMArena Math | 1241 | 1475 |
| GSM8K | 94.5% | — |

## Knowledge

- Deepseek Coder v2: 31.7 (#209)
- Hy3: 40.2 (#108)

| Benchmark | Deepseek Coder v2 | Hy3 |
|---|---|---|
| LMArena Expert | 1181 | 1461 |
| ARC (AI2) Challenge | 64.3% | — |

## Multilingual

- Deepseek Coder v2: 36.4 (#235)
- Hy3: 53.7 (#60)

| Benchmark | Deepseek Coder v2 | Hy3 |
|---|---|---|
| LMArena Non-English | 1182 | 1427 |
| LMArena Chinese | 1201 | 1496 |
| LMArena French | 1184 | 1461 |
| LMArena Korean | 1104 | 1390 |
| LMArena Russian | 1188 | 1435 |
| LMArena Spanish | 1153 | 1448 |
| LMArena German | 1164 | — |
| LMArena Japanese | 1126 | — |

## Instruction Following

- Deepseek Coder v2: 61.9 (#237)
- Hy3: 75.3 (#68)

| Benchmark | Deepseek Coder v2 | Hy3 |
|---|---|---|
| LMArena Instruction Following | 1180 | 1425 |

## Long Context

- Deepseek Coder v2: 37.0 (#219)
- Hy3: 44.1 (#70)

| Benchmark | Deepseek Coder v2 | Hy3 |
|---|---|---|
| LMArena Longer Query | 1219 | 1443 |

## Writing & Preference

- Deepseek Coder v2: 38.4 (#249)
- Hy3: 62.5 (#75)

| Benchmark | Deepseek Coder v2 | Hy3 |
|---|---|---|
| LMArena Text | 1191 | 1442 |
| LMArena Creative Writing | 1120 | 1399 |
| LMArena Multi-Turn | 1176 | 1436 |

## FAQ

### Is Deepseek Coder v2 better than Hy3?

Hy3 is the stronger model overall, scoring 43.6 to 35.3 on the Noometry Index.

### Is Deepseek Coder v2 or Hy3 better for coding?

Hy3 scores higher on coding benchmarks: 46.6 versus 37.5 in the Noometry coding category.

### How many benchmarks do Deepseek Coder v2 and Hy3 share?

15 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and Hy3 has 17.