# gpt-oss-20b vs Qwen3.5-9B

> Qwen3.5-9B is the stronger model overall, scoring 33.8 to 32.5 on the Noometry Index. gpt-oss-20b costs 3.1× less per token, which makes it the better buy when Qwen3.5-9B's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gpt-oss-20b-vs-qwen3-5-9b
- Last updated: 2026-10-10
- Shared benchmarks: 9

## Summary

- They share 9 benchmarks with published results for both. gpt-oss-20b scores higher in 2 categories and Qwen3.5-9B in 3 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3.5-9B leads 46.0 to 34.6.
- The biggest single-benchmark swing is GPQA Diamond: 60.8% for gpt-oss-20b and 79% for Qwen3.5-9B.
- gpt-oss-20b is cheaper at $0.018 / $0.09 per million input/output tokens, against $0.10 / $0.15 for Qwen3.5-9B.
- Qwen3.5-9B accepts more context: 262K tokens versus 131K.

## Snapshot

| | gpt-oss-20b | Qwen3.5-9B |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 32.5 | 33.8 |
| Rank | 255 | 236 |
| Context | 131K | 262K |
| Input $/M | $0.018 | $0.10 |
| Output $/M | $0.09 | $0.15 |
| Weights | Open | Open |

## Coding

- gpt-oss-20b: 37.6 (#192)
- Qwen3.5-9B: 35.9 (#217)

| Benchmark | gpt-oss-20b | Qwen3.5-9B |
|---|---|---|
| SciCode | 34.4% | 27.5% |
| WeirdML | 40.9% | — |
| LMArena Coding | 1306 | — |
| ALE-Bench | 566.05 | — |

## Agentic & Tool Use

- gpt-oss-20b: 9.3 (#154)
- Qwen3.5-9B: 14.5 (#151)

| Benchmark | gpt-oss-20b | Qwen3.5-9B |
|---|---|---|
| Terminal-Bench | 3.4% | 9.2% |

## Reasoning

- gpt-oss-20b: 19.3 (#261)
- Qwen3.5-9B: 23.1 (#182)

| Benchmark | gpt-oss-20b | Qwen3.5-9B |
|---|---|---|
| CritPt | 1.4% | 0.3% |
| Chess Puzzles | 4% | 12% |
| DTBench | 68% | 71.2% |
| LMCA | 14.5% | 24.5% |
| Epoch Capabilities Index | 137.82 | 139.46 |
| Kagi LLM Benchmark | 53.2% | — |
| LMArena Hard Prompts | 1274 | — |

## Math

- gpt-oss-20b: 39.4 (#103)
- Qwen3.5-9B: 34.8 (#192)

| Benchmark | gpt-oss-20b | Qwen3.5-9B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 65.3% | 61.7% |
| MathArena Final-Answer Competitions | — | 48.5% |
| Omni-MATH | 56.5% | — |
| LMArena Math | 1317 | — |

## Knowledge

- gpt-oss-20b: 34.6 (#195)
- Qwen3.5-9B: 46.0 (#84)

| Benchmark | gpt-oss-20b | Qwen3.5-9B |
|---|---|---|
| GPQA Diamond | 60.8% | 79% |
| MMLU-Pro | 74% | — |
| GPQA (HELM) | 59.4% | — |
| LMArena Expert | 1258 | — |

## Multilingual

- gpt-oss-20b: 42.2 (#197)
- Qwen3.5-9B: —

| Benchmark | gpt-oss-20b | Qwen3.5-9B |
|---|---|---|
| LMArena Non-English | 1268 | — |
| LMArena Chinese | 1314 | — |
| LMArena German | 1255 | — |
| LMArena Japanese | 1244 | — |
| LMArena Korean | 1236 | — |
| LMArena Russian | 1278 | — |
| LMArena Spanish | 1267 | — |

## Instruction Following

- gpt-oss-20b: 61.8 (#240)
- Qwen3.5-9B: —

| Benchmark | gpt-oss-20b | Qwen3.5-9B |
|---|---|---|
| IFEval | 73.2% | — |
| LMArena Instruction Following | 1236 | — |

## Long Context

- gpt-oss-20b: 37.9 (#209)
- Qwen3.5-9B: —

| Benchmark | gpt-oss-20b | Qwen3.5-9B |
|---|---|---|
| LMArena Longer Query | 1250 | — |

## Writing & Preference

- gpt-oss-20b: 35.5 (#265)
- Qwen3.5-9B: —

| Benchmark | gpt-oss-20b | Qwen3.5-9B |
|---|---|---|
| LMArena Text | 1287 | — |
| LMArena Creative Writing | 1201 | — |
| EQ-Bench Creative Writing | 666 | — |
| WildBench | 73.7% | — |
| LMArena Multi-Turn | 1268 | — |

## FAQ

### Is gpt-oss-20b better than Qwen3.5-9B?

Qwen3.5-9B is the stronger model overall, scoring 33.8 to 32.5 on the Noometry Index. gpt-oss-20b costs 3.1× less per token, which makes it the better buy when Qwen3.5-9B's lead doesn't matter for your workload.

### Which is cheaper, gpt-oss-20b or Qwen3.5-9B?

gpt-oss-20b is cheaper. It lists at $0.018 per million input tokens and $0.09 per million output tokens; Qwen3.5-9B lists at $0.10 and $0.15.

### Is gpt-oss-20b or Qwen3.5-9B better for coding?

gpt-oss-20b scores higher on coding benchmarks: 37.6 versus 35.9 in the Noometry coding category.

### Which has the bigger context window?

Qwen3.5-9B does, with 262K tokens against 131K.

### How many benchmarks do gpt-oss-20b and Qwen3.5-9B share?

9 benchmarks have published results for both models. gpt-oss-20b has 34 scored results on Noometry and Qwen3.5-9B has 10.