# Llama 3.1-405B vs Longcat Flash Chat

> Longcat Flash Chat is the stronger model overall, scoring 41.5 to 30.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/llama-3-1-405b-vs-longcat-flash-chat
- Last updated: 2026-10-10
- Shared benchmarks: 18

## Summary

- They share 18 benchmarks with published results for both. Llama 3.1-405B scores higher in 0 categories and Longcat Flash Chat in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Longcat Flash Chat leads 61.2 to 39.1.

## Snapshot

| | Llama 3.1-405B | Longcat Flash Chat |
|---|---|---|
| Provider | Meta | Meituan |
| Noometry Index | 30.4 | 41.5 |
| Rank | 279 | 121 |
| Context | — | — |
| Input $/M | — | — |
| Output $/M | — | — |
| Weights | Open | Open |

## Coding

- Llama 3.1-405B: 32.7 (#257)
- Longcat Flash Chat: 42.9 (#85)

| Benchmark | Llama 3.1-405B | Longcat Flash Chat |
|---|---|---|
| LMArena Coding | 1291 | 1471 |
| WeirdML | 21.4% | — |

## Agentic & Tool Use

- Llama 3.1-405B: 20.9 (#131)
- Longcat Flash Chat: —

| Benchmark | Llama 3.1-405B | Longcat Flash Chat |
|---|---|---|
| TheAgentCompany | 7.4% | — |
| Cybench | 7.5% | — |

## Reasoning

- Llama 3.1-405B: 16.4 (#293)
- Longcat Flash Chat: 18.3 (#267)

| Benchmark | Llama 3.1-405B | Longcat Flash Chat |
|---|---|---|
| Kagi LLM Benchmark | 45% | 43.9% |
| LMArena Hard Prompts | 1270 | 1440 |
| SimpleBench | 23% | — |
| NYT Connections (extended) | — | 17.7% |
| DTBench | 61.4% | — |
| BIG-Bench Hard | 82.9% | — |
| Epoch Capabilities Index | 128.76 | — |
| ForecastBench | 59.9 | — |
| HellaSwag | 89.2% | — |
| PIQA | 85.9% | — |
| WinoGrande | 89.2% | — |

## Math

- Llama 3.1-405B: 17.3 (#283)
- Longcat Flash Chat: 36.6 (#107)

| Benchmark | Llama 3.1-405B | Longcat Flash Chat |
|---|---|---|
| LMArena Math | 1281 | 1441 |
| OTIS Mock AIME 2024-2025 | 9.7% | — |
| Omni-MATH | 24.9% | — |
| MATH Level 5 | 49.8% | — |

## Knowledge

- Llama 3.1-405B: 30.4 (#221)
- Longcat Flash Chat: 40.0 (#111)

| Benchmark | Llama 3.1-405B | Longcat Flash Chat |
|---|---|---|
| LMArena Expert | 1243 | 1454 |
| GPQA Diamond | 50.9% | — |
| MMLU-Pro | 72.3% | — |
| Confabulations | 17.6% | — |
| GPQA (HELM) | 52.2% | — |
| ARC (AI2) Challenge | 95.3% | — |
| MMLU | 84.5% | — |
| TriviaQA | 82.7% | — |

## Multilingual

- Llama 3.1-405B: 40.9 (#209)
- Longcat Flash Chat: 52.1 (#95)

| Benchmark | Llama 3.1-405B | Longcat Flash Chat |
|---|---|---|
| LMArena Non-English | 1248 | 1404 |
| LMArena Chinese | 1242 | 1465 |
| LMArena French | 1278 | 1457 |
| LMArena German | 1252 | 1409 |
| LMArena Japanese | 1208 | 1373 |
| LMArena Korean | 1184 | 1372 |
| LMArena Russian | 1265 | 1396 |
| LMArena Spanish | 1259 | 1445 |

## Instruction Following

- Llama 3.1-405B: 66.1 (#209)
- Longcat Flash Chat: 74.6 (#92)

| Benchmark | Llama 3.1-405B | Longcat Flash Chat |
|---|---|---|
| LMArena Instruction Following | 1259 | 1411 |
| IFEval | 81.1% | — |

## Long Context

- Llama 3.1-405B: 38.4 (#192)
- Longcat Flash Chat: 43.5 (#89)

| Benchmark | Llama 3.1-405B | Longcat Flash Chat |
|---|---|---|
| LMArena Longer Query | 1266 | 1425 |

## Writing & Preference

- Llama 3.1-405B: 39.1 (#246)
- Longcat Flash Chat: 61.2 (#88)

| Benchmark | Llama 3.1-405B | Longcat Flash Chat |
|---|---|---|
| LMArena Text | 1284 | 1427 |
| LMArena Creative Writing | 1262 | 1388 |
| LMArena Multi-Turn | 1297 | 1418 |
| EQ-Bench Creative Writing | 870 | — |
| WildBench | 78.3% | — |

## FAQ

### Is Llama 3.1-405B better than Longcat Flash Chat?

Longcat Flash Chat is the stronger model overall, scoring 41.5 to 30.4 on the Noometry Index.

### Is Llama 3.1-405B or Longcat Flash Chat better for coding?

Longcat Flash Chat scores higher on coding benchmarks: 42.9 versus 32.7 in the Noometry coding category.

### How many benchmarks do Llama 3.1-405B and Longcat Flash Chat share?

18 benchmarks have published results for both models. Llama 3.1-405B has 42 scored results on Noometry and Longcat Flash Chat has 19.