# gpt-oss-20b vs Hy3

> Hy3 is the stronger model overall, scoring 44.2 to 32.5 on the Noometry Index. gpt-oss-20b costs 6.4× less per token, which makes it the better buy when Hy3's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gpt-oss-20b-vs-hy3
- Last updated: 2026-10-11
- Shared benchmarks: 16

## Summary

- They share 16 benchmarks with published results for both. gpt-oss-20b scores higher in 0 categories and Hy3 in 8 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Hy3 leads 62.2 to 35.5.
- gpt-oss-20b is cheaper at $0.018 / $0.09 per million input/output tokens, against $0.13 / $0.53 for Hy3.
- Hy3 accepts more context: 262K tokens versus 131K.

## Snapshot

| | gpt-oss-20b | Hy3 |
|---|---|---|
| Provider | OpenAI | Tencent |
| Noometry Index | 32.5 | 44.2 |
| Rank | 255 | 79 |
| Context | 131K | 262K |
| Input $/M | $0.018 | $0.13 |
| Output $/M | $0.09 | $0.53 |
| Weights | Open | Open |

## Coding

- gpt-oss-20b: 37.6 (#192)
- Hy3: 46.8 (#63)

| Benchmark | gpt-oss-20b | Hy3 |
|---|---|---|
| LMArena Coding | 1306 | 1464 |
| LMArena WebDev | — | 1508 |
| SciCode | 34.4% | — |
| WeirdML | 40.9% | — |
| ALE-Bench | 566.05 | — |

## Agentic & Tool Use

- gpt-oss-20b: 9.3 (#154)
- Hy3: —

| Benchmark | gpt-oss-20b | Hy3 |
|---|---|---|
| Terminal-Bench | 3.4% | — |

## Reasoning

- gpt-oss-20b: 19.3 (#261)
- Hy3: 26.1 (#136)

| Benchmark | gpt-oss-20b | Hy3 |
|---|---|---|
| LMArena Hard Prompts | 1274 | 1447 |
| Kagi LLM Benchmark | 53.2% | — |
| NYT Connections (extended) | — | 41.2% |
| CritPt | 1.4% | — |
| Chess Puzzles | 4% | — |
| DTBench | 68% | — |
| LMCA | 14.5% | — |
| Epoch Capabilities Index | 137.82 | — |

## Math

- gpt-oss-20b: 39.4 (#103)
- Hy3: 40.1 (#93)

| Benchmark | gpt-oss-20b | Hy3 |
|---|---|---|
| LMArena Math | 1317 | 1475 |
| OTIS Mock AIME 2024-2025 | 65.3% | — |
| Omni-MATH | 56.5% | — |

## Knowledge

- gpt-oss-20b: 34.6 (#195)
- Hy3: 40.8 (#114)

| Benchmark | gpt-oss-20b | Hy3 |
|---|---|---|
| LMArena Expert | 1258 | 1460 |
| GPQA Diamond | 60.8% | — |
| MMLU-Pro | 74% | — |
| GPQA (HELM) | 59.4% | — |

## Multilingual

- gpt-oss-20b: 42.2 (#197)
- Hy3: 53.5 (#65)

| Benchmark | gpt-oss-20b | Hy3 |
|---|---|---|
| LMArena Non-English | 1268 | 1426 |
| LMArena Chinese | 1314 | 1493 |
| LMArena German | 1255 | 1439 |
| LMArena Japanese | 1244 | 1392 |
| LMArena Korean | 1236 | 1395 |
| LMArena Russian | 1278 | 1432 |
| LMArena Spanish | 1267 | 1456 |
| LMArena French | — | 1461 |

## Instruction Following

- gpt-oss-20b: 61.8 (#240)
- Hy3: 75.1 (#70)

| Benchmark | gpt-oss-20b | Hy3 |
|---|---|---|
| LMArena Instruction Following | 1236 | 1426 |
| IFEval | 73.2% | — |

## Long Context

- gpt-oss-20b: 37.9 (#209)
- Hy3: 44.1 (#75)

| Benchmark | gpt-oss-20b | Hy3 |
|---|---|---|
| LMArena Longer Query | 1250 | 1442 |

## Writing & Preference

- gpt-oss-20b: 35.5 (#265)
- Hy3: 62.2 (#81)

| Benchmark | gpt-oss-20b | Hy3 |
|---|---|---|
| LMArena Text | 1287 | 1439 |
| LMArena Creative Writing | 1201 | 1402 |
| LMArena Multi-Turn | 1268 | 1436 |
| EQ-Bench Creative Writing | 666 | — |
| WildBench | 73.7% | — |

## FAQ

### Is gpt-oss-20b better than Hy3?

Hy3 is the stronger model overall, scoring 44.2 to 32.5 on the Noometry Index. gpt-oss-20b costs 6.4× less per token, which makes it the better buy when Hy3's lead doesn't matter for your workload.

### Which is cheaper, gpt-oss-20b or Hy3?

gpt-oss-20b is cheaper. It lists at $0.018 per million input tokens and $0.09 per million output tokens; Hy3 lists at $0.13 and $0.53.

### Is gpt-oss-20b or Hy3 better for coding?

Hy3 scores higher on coding benchmarks: 46.8 versus 37.6 in the Noometry coding category.

### Which has the bigger context window?

Hy3 does, with 262K tokens against 131K.

### How many benchmarks do gpt-oss-20b and Hy3 share?

16 benchmarks have published results for both models. gpt-oss-20b has 34 scored results on Noometry and Hy3 has 19.