# GLM-5.3 vs Llama 3.1-405B

> GLM-5.3 is the stronger model overall, scoring 54.8 to 30.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/glm-5-3-vs-llama-3-1-405b
- Last updated: 2026-10-10
- Shared benchmarks: 23

## Summary

- They share 23 benchmarks with published results for both. GLM-5.3 scores higher in 9 categories and Llama 3.1-405B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GLM-5.3 leads 62.3 to 18.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 91.1% for GLM-5.3 and 9.7% for Llama 3.1-405B.

## Snapshot

| | GLM-5.3 | Llama 3.1-405B |
|---|---|---|
| Provider | Z.ai (Zhipu) | Meta |
| Noometry Index | 54.8 | 30.7 |
| Rank | 26 | 288 |
| Context | 1M | — |
| Input $/M | $1.40 | — |
| Output $/M | $4.40 | — |
| Weights | Open | Open |

## Coding

- GLM-5.3: 59.5 (#14)
- Llama 3.1-405B: 33.1 (#262)

| Benchmark | GLM-5.3 | Llama 3.1-405B |
|---|---|---|
| WeirdML | 75.4% | 21.4% |
| LMArena Coding | 1496 | 1291 |
| DeepSWE | 69% | — |
| FrontierCode | 40.1% | — |
| CursorBench | 42.6% | — |
| LMArena WebDev | 1622 | — |
| FrontierSWE | 30.2% | — |
| SciCode | 59% | — |
| ALE-Bench | 1,317 | — |

## Agentic & Tool Use

- GLM-5.3: 36.4 (#38)
- Llama 3.1-405B: 21.0 (#140)

| Benchmark | GLM-5.3 | Llama 3.1-405B |
|---|---|---|
| APEX-Agents | 56.6% | — |
| TheAgentCompany | — | 7.4% |
| Cybench | — | 7.5% |
| Vending-Bench 2 | 8,164 | — |

## Reasoning

- GLM-5.3: 46.1 (#46)
- Llama 3.1-405B: 16.8 (#300)

| Benchmark | GLM-5.3 | Llama 3.1-405B |
|---|---|---|
| LMArena Hard Prompts | 1489 | 1269 |
| DTBench | 87.7% | 61.4% |
| Epoch Capabilities Index | 155.61 | 128.75 |
| SimpleBench | — | 23% |
| Kagi LLM Benchmark | — | 45% |
| NYT Connections (extended) | 74.2% | — |
| CritPt | 19.1% | — |
| Chess Puzzles | 21% | — |
| Mystery Game Puzzles | 33% | — |
| LMCA | 55.5% | — |
| Bench to the Future 3 | 0.15 | — |
| BIG-Bench Hard | — | 82.9% |
| ForecastBench | — | 59.9 |
| HellaSwag | — | 89.2% |
| PIQA | — | 85.9% |
| WinoGrande | — | 89.2% |

## Math

- GLM-5.3: 62.3 (#33)
- Llama 3.1-405B: 18.4 (#290)

| Benchmark | GLM-5.3 | Llama 3.1-405B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 91.1% | 9.7% |
| LMArena Math | 1489 | 1281 |
| FrontierMath (Tiers 1-3) | 68.8% | — |
| FrontierMath Tier 4 | 29.3% | — |
| ProofBench | 49% | — |
| Omni-MATH | — | 24.9% |
| MATH Level 5 | — | 49.8% |

## Knowledge

- GLM-5.3: 58.3 (#37)
- Llama 3.1-405B: 30.4 (#227)

| Benchmark | GLM-5.3 | Llama 3.1-405B |
|---|---|---|
| GPQA Diamond | 90.9% | 50.9% |
| LMArena Expert | 1516 | 1243 |
| SimpleQA Verified | 41% | — |
| MMLU-Pro | — | 72.3% |
| Confabulations | — | 17.6% |
| GPQA (HELM) | — | 52.2% |
| ARC (AI2) Challenge | — | 95.3% |
| MMLU | — | 84.5% |
| TriviaQA | — | 82.7% |

## Multilingual

- GLM-5.3: 55.7 (#28)
- Llama 3.1-405B: 40.7 (#214)

| Benchmark | GLM-5.3 | Llama 3.1-405B |
|---|---|---|
| LMArena Non-English | 1457 | 1248 |
| LMArena Chinese | 1528 | 1242 |
| LMArena French | 1499 | 1279 |
| LMArena German | 1499 | 1252 |
| LMArena Japanese | 1453 | 1208 |
| LMArena Korean | 1472 | 1184 |
| LMArena Russian | 1463 | 1265 |
| LMArena Spanish | 1460 | 1260 |

## Instruction Following

- GLM-5.3: 77.5 (#23)
- Llama 3.1-405B: 65.9 (#214)

| Benchmark | GLM-5.3 | Llama 3.1-405B |
|---|---|---|
| LMArena Instruction Following | 1477 | 1259 |
| IFEval | — | 81.1% |

## Long Context

- GLM-5.3: 45.4 (#41)
- Llama 3.1-405B: 38.4 (#197)

| Benchmark | GLM-5.3 | Llama 3.1-405B |
|---|---|---|
| LMArena Longer Query | 1482 | 1266 |

## Writing & Preference

- GLM-5.3: 75.7 (#6)
- Llama 3.1-405B: 38.9 (#251)

| Benchmark | GLM-5.3 | Llama 3.1-405B |
|---|---|---|
| LMArena Text | 1471 | 1284 |
| LMArena Creative Writing | 1457 | 1262 |
| EQ-Bench Creative Writing | 2075 | 870 |
| LMArena Multi-Turn | 1472 | 1297 |
| WildBench | — | 78.3% |

## FAQ

### Is GLM-5.3 better than Llama 3.1-405B?

GLM-5.3 is the stronger model overall, scoring 54.8 to 30.7 on the Noometry Index.

### Is GLM-5.3 or Llama 3.1-405B better for coding?

GLM-5.3 scores higher on coding benchmarks: 59.5 versus 33.1 in the Noometry coding category.

### How many benchmarks do GLM-5.3 and Llama 3.1-405B share?

23 benchmarks have published results for both models. GLM-5.3 has 42 scored results on Noometry and Llama 3.1-405B has 42.