# GLM-5.3-Flash vs gpt-oss-20b

> GLM-5.3-Flash is the stronger model overall, scoring 49.6 to 32.1 on the Noometry Index. gpt-oss-20b costs 6.6× less per token, which makes it the better buy when GLM-5.3-Flash's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/glm-5-3-flash-vs-gpt-oss-20b
- Last updated: 2026-10-10
- Shared benchmarks: 23

## Summary

- They share 23 benchmarks with published results for both. GLM-5.3-Flash scores higher in 9 categories and gpt-oss-20b in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GLM-5.3-Flash leads 65.4 to 35.8.
- The biggest single-benchmark swing is GPQA Diamond: 90.2% for GLM-5.3-Flash and 60.8% for gpt-oss-20b.
- gpt-oss-20b is cheaper at $0.018 / $0.09 per million input/output tokens, against $0.15 / $0.50 for GLM-5.3-Flash.
- GLM-5.3-Flash accepts more context: 1M tokens versus 131K.

## Snapshot

| | GLM-5.3-Flash | gpt-oss-20b |
|---|---|---|
| Provider | Z.ai (Zhipu) | OpenAI |
| Noometry Index | 49.6 | 32.1 |
| Rank | 45 | 249 |
| Context | 1M | 131K |
| Input $/M | $0.15 | $0.018 |
| Output $/M | $0.50 | $0.09 |
| Weights | Open | Open |

## Coding

- GLM-5.3-Flash: 53.7 (#26)
- gpt-oss-20b: 37.1 (#185)

| Benchmark | GLM-5.3-Flash | gpt-oss-20b |
|---|---|---|
| SciCode | 51.6% | 34.4% |
| LMArena Coding | 1506 | 1307 |
| ALE-Bench | 303.55 | 566.05 |
| DeepSWE | 63.4% | — |
| FrontierCode | 31.8% | — |
| LMArena WebDev | 1614 | — |
| WeirdML | — | 40.9% |

## Agentic & Tool Use

- GLM-5.3-Flash: 34.3 (#41)
- gpt-oss-20b: 9.2 (#146)

| Benchmark | GLM-5.3-Flash | gpt-oss-20b |
|---|---|---|
| Terminal-Bench | — | 3.4% |
| APEX-Agents | 52.8% | — |
| GDP.pdf | 14% | — |

## Reasoning

- GLM-5.3-Flash: 31.7 (#69)
- gpt-oss-20b: 18.8 (#251)

| Benchmark | GLM-5.3-Flash | gpt-oss-20b |
|---|---|---|
| CritPt | 15.4% | 1.4% |
| Chess Puzzles | 14% | 4% |
| LMArena Hard Prompts | 1490 | 1274 |
| Epoch Capabilities Index | 151.9 | 137.81 |
| Kagi LLM Benchmark | — | 53.2% |
| Mystery Game Puzzles | 8% | — |
| DTBench | — | 68% |
| LMCA | — | 14.5% |
| Surface Evolver Bench | 52.5% | — |
| Bench to the Future 3 | 0.15 | — |

## Math

- GLM-5.3-Flash: 53.1 (#38)
- gpt-oss-20b: 37.6 (#93)

| Benchmark | GLM-5.3-Flash | gpt-oss-20b |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 93.9% | 65.3% |
| LMArena Math | 1499 | 1317 |
| FrontierMath (Tiers 1-3) | 55.8% | — |
| FrontierMath Tier 4 | 17.1% | — |
| ProofBench | 21% | — |
| Omni-MATH | — | 56.5% |

## Knowledge

- GLM-5.3-Flash: 58.4 (#33)
- gpt-oss-20b: 34.6 (#186)

| Benchmark | GLM-5.3-Flash | gpt-oss-20b |
|---|---|---|
| GPQA Diamond | 90.2% | 60.8% |
| LMArena Expert | 1512 | 1258 |
| MMLU-Pro | — | 74% |
| GPQA (HELM) | — | 59.4% |

## Multimodal

- GLM-5.3-Flash: 42.6 (#25)
- gpt-oss-20b: —

| Benchmark | GLM-5.3-Flash | gpt-oss-20b |
|---|---|---|
| LMArena Vision | 1304 | — |

## Multilingual

- GLM-5.3-Flash: 56.2 (#24)
- gpt-oss-20b: 42.4 (#192)

| Benchmark | GLM-5.3-Flash | gpt-oss-20b |
|---|---|---|
| LMArena Non-English | 1462 | 1269 |
| LMArena Chinese | 1529 | 1311 |
| LMArena German | 1457 | 1249 |
| LMArena Japanese | 1420 | 1245 |
| LMArena Korean | 1443 | 1239 |
| LMArena Russian | 1472 | 1279 |
| LMArena Spanish | 1471 | 1267 |
| LMArena French | 1495 | — |

## Instruction Following

- GLM-5.3-Flash: 77.8 (#17)
- gpt-oss-20b: 62.0 (#235)

| Benchmark | GLM-5.3-Flash | gpt-oss-20b |
|---|---|---|
| LMArena Instruction Following | 1480 | 1236 |
| IFEval | — | 73.2% |

## Long Context

- GLM-5.3-Flash: 45.3 (#41)
- gpt-oss-20b: 37.9 (#203)

| Benchmark | GLM-5.3-Flash | gpt-oss-20b |
|---|---|---|
| LMArena Longer Query | 1479 | 1251 |

## Writing & Preference

- GLM-5.3-Flash: 65.4 (#49)
- gpt-oss-20b: 35.8 (#260)

| Benchmark | GLM-5.3-Flash | gpt-oss-20b |
|---|---|---|
| LMArena Text | 1471 | 1288 |
| LMArena Creative Writing | 1440 | 1201 |
| LMArena Multi-Turn | 1471 | 1268 |
| EQ-Bench Creative Writing | — | 666 |
| WildBench | — | 73.7% |

## FAQ

### Is GLM-5.3-Flash better than gpt-oss-20b?

GLM-5.3-Flash is the stronger model overall, scoring 49.6 to 32.1 on the Noometry Index. gpt-oss-20b costs 6.6× less per token, which makes it the better buy when GLM-5.3-Flash's lead doesn't matter for your workload.

### Which is cheaper, GLM-5.3-Flash or gpt-oss-20b?

gpt-oss-20b is cheaper. It lists at $0.018 per million input tokens and $0.09 per million output tokens; GLM-5.3-Flash lists at $0.15 and $0.50.

### Is GLM-5.3-Flash or gpt-oss-20b better for coding?

GLM-5.3-Flash scores higher on coding benchmarks: 53.7 versus 37.1 in the Noometry coding category.

### Which has the bigger context window?

GLM-5.3-Flash does, with 1M tokens against 131K.

### How many benchmarks do GLM-5.3-Flash and gpt-oss-20b share?

23 benchmarks have published results for both models. GLM-5.3-Flash has 36 scored results on Noometry and gpt-oss-20b has 34.