# gpt-oss-20b vs Longcat Flash Chat

> Longcat Flash Chat is the stronger model overall, scoring 42.1 to 32.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-oss-20b-vs-longcat-flash-chat
- Last updated: 2026-10-10
- Shared benchmarks: 17

## Summary

- They share 17 benchmarks with published results for both. gpt-oss-20b scores higher in 2 categories and Longcat Flash Chat in 6 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Longcat Flash Chat leads 61.0 to 35.5.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 53.2% for gpt-oss-20b and 43.9% for Longcat Flash Chat.

## Snapshot

| | gpt-oss-20b | Longcat Flash Chat |
|---|---|---|
| Provider | OpenAI | Meituan |
| Noometry Index | 32.5 | 42.1 |
| Rank | 255 | 120 |
| Context | 131K | — |
| Input $/M | $0.018 | — |
| Output $/M | $0.09 | — |
| Weights | Open | Open |

## Coding

- gpt-oss-20b: 37.6 (#192)
- Longcat Flash Chat: 43.5 (#87)

| Benchmark | gpt-oss-20b | Longcat Flash Chat |
|---|---|---|
| LMArena Coding | 1306 | 1471 |
| SciCode | 34.4% | — |
| WeirdML | 40.9% | — |
| ALE-Bench | 566.05 | — |

## Agentic & Tool Use

- gpt-oss-20b: 9.3 (#154)
- Longcat Flash Chat: —

| Benchmark | gpt-oss-20b | Longcat Flash Chat |
|---|---|---|
| Terminal-Bench | 3.4% | — |

## Reasoning

- gpt-oss-20b: 19.3 (#261)
- Longcat Flash Chat: 19.0 (#272)

| Benchmark | gpt-oss-20b | Longcat Flash Chat |
|---|---|---|
| Kagi LLM Benchmark | 53.2% | 43.9% |
| LMArena Hard Prompts | 1274 | 1440 |
| NYT Connections (extended) | — | 17.7% |
| CritPt | 1.4% | — |
| Chess Puzzles | 4% | — |
| DTBench | 68% | — |
| LMCA | 14.5% | — |
| Epoch Capabilities Index | 137.82 | — |

## Math

- gpt-oss-20b: 39.4 (#103)
- Longcat Flash Chat: 39.4 (#107)

| Benchmark | gpt-oss-20b | Longcat Flash Chat |
|---|---|---|
| LMArena Math | 1317 | 1442 |
| OTIS Mock AIME 2024-2025 | 65.3% | — |
| Omni-MATH | 56.5% | — |

## Knowledge

- gpt-oss-20b: 34.6 (#195)
- Longcat Flash Chat: 40.6 (#116)

| Benchmark | gpt-oss-20b | Longcat Flash Chat |
|---|---|---|
| LMArena Expert | 1258 | 1454 |
| GPQA Diamond | 60.8% | — |
| MMLU-Pro | 74% | — |
| GPQA (HELM) | 59.4% | — |

## Multilingual

- gpt-oss-20b: 42.2 (#197)
- Longcat Flash Chat: 51.9 (#101)

| Benchmark | gpt-oss-20b | Longcat Flash Chat |
|---|---|---|
| LMArena Non-English | 1268 | 1404 |
| LMArena Chinese | 1314 | 1465 |
| LMArena German | 1255 | 1408 |
| LMArena Japanese | 1244 | 1373 |
| LMArena Korean | 1236 | 1371 |
| LMArena Russian | 1278 | 1395 |
| LMArena Spanish | 1267 | 1445 |
| LMArena French | — | 1456 |

## Instruction Following

- gpt-oss-20b: 61.8 (#240)
- Longcat Flash Chat: 74.4 (#96)

| Benchmark | gpt-oss-20b | Longcat Flash Chat |
|---|---|---|
| LMArena Instruction Following | 1236 | 1411 |
| IFEval | 73.2% | — |

## Long Context

- gpt-oss-20b: 37.9 (#209)
- Longcat Flash Chat: 43.5 (#93)

| Benchmark | gpt-oss-20b | Longcat Flash Chat |
|---|---|---|
| LMArena Longer Query | 1250 | 1425 |

## Writing & Preference

- gpt-oss-20b: 35.5 (#265)
- Longcat Flash Chat: 61.0 (#91)

| Benchmark | gpt-oss-20b | Longcat Flash Chat |
|---|---|---|
| LMArena Text | 1287 | 1427 |
| LMArena Creative Writing | 1201 | 1388 |
| LMArena Multi-Turn | 1268 | 1418 |
| EQ-Bench Creative Writing | 666 | — |
| WildBench | 73.7% | — |

## FAQ

### Is gpt-oss-20b better than Longcat Flash Chat?

Longcat Flash Chat is the stronger model overall, scoring 42.1 to 32.5 on the Noometry Index.

### Is gpt-oss-20b or Longcat Flash Chat better for coding?

Longcat Flash Chat scores higher on coding benchmarks: 43.5 versus 37.6 in the Noometry coding category.

### How many benchmarks do gpt-oss-20b and Longcat Flash Chat share?

17 benchmarks have published results for both models. gpt-oss-20b has 34 scored results on Noometry and Longcat Flash Chat has 19.