# GLM-5.3 vs Grok 4.3

> GLM-5.3 is the stronger model overall, scoring 54.8 to 43.8 on the Noometry Index.

- Canonical page: https://noometry.com/compare/glm-5-3-vs-grok-4-3
- Last updated: 2026-10-11
- Shared benchmarks: 34

## Summary

- They share 34 benchmarks with published results for both. GLM-5.3 scores higher in 9 categories and Grok 4.3 in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in coding, where GLM-5.3 leads 59.5 to 41.6.
- The biggest single-benchmark swing is ProofBench: 49% for GLM-5.3 and 11% for Grok 4.3.
- Grok 4.3 is cheaper at $1.25 / $2.50 per million input/output tokens, against $1.40 / $4.40 for GLM-5.3.
- GLM-5.3 has downloadable open weights; the other is API-only.

## Snapshot

| | GLM-5.3 | Grok 4.3 |
|---|---|---|
| Provider | Z.ai (Zhipu) | xAI |
| Noometry Index | 54.8 | 43.8 |
| Rank | 26 | 86 |
| Context | 1M | 1M |
| Input $/M | $1.40 | $1.25 |
| Output $/M | $4.40 | $2.50 |
| Weights | Open | Proprietary |

## Coding

- GLM-5.3: 59.5 (#14)
- Grok 4.3: 41.6 (#121)

| Benchmark | GLM-5.3 | Grok 4.3 |
|---|---|---|
| LMArena WebDev | 1622 | 1357 |
| SciCode | 59% | 47.3% |
| WeirdML | 75.4% | 49.9% |
| LMArena Coding | 1496 | 1415 |
| ALE-Bench | 1,317 | 944.17 |
| DeepSWE | 69% | — |
| FrontierCode | 40.1% | — |
| CursorBench | 42.6% | — |
| FrontierSWE | 30.2% | — |

## Agentic & Tool Use

- GLM-5.3: 36.4 (#38)
- Grok 4.3: 27.7 (#99)

| Benchmark | GLM-5.3 | Grok 4.3 |
|---|---|---|
| Vending-Bench 2 | 8,164 | 35.26 |
| APEX-Agents | 56.6% | — |
| GDP.pdf | — | 8% |
| LMArena Search | — | 1165 |

## Reasoning

- GLM-5.3: 46.1 (#46)
- Grok 4.3: 35.9 (#68)

| Benchmark | GLM-5.3 | Grok 4.3 |
|---|---|---|
| NYT Connections (extended) | 74.2% | 55.2% |
| CritPt | 19.1% | 8% |
| Chess Puzzles | 21% | 25% |
| LMArena Hard Prompts | 1489 | 1396 |
| DTBench | 87.7% | 90.7% |
| LMCA | 55.5% | 38.3% |
| Epoch Capabilities Index | 155.61 | 149.16 |
| Mystery Game Puzzles | 33% | — |
| Bench to the Future 3 | 0.15 | — |
| ForecastBench | — | 60.3 |

## Math

- GLM-5.3: 62.3 (#33)
- Grok 4.3: 46.0 (#74)

| Benchmark | GLM-5.3 | Grok 4.3 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 68.8% | 42.8% |
| FrontierMath Tier 4 | 29.3% | 14.6% |
| OTIS Mock AIME 2024-2025 | 91.1% | 93.3% |
| ProofBench | 49% | 11% |
| LMArena Math | 1489 | 1388 |

## Knowledge

- GLM-5.3: 58.3 (#37)
- Grok 4.3: 52.5 (#62)

| Benchmark | GLM-5.3 | Grok 4.3 |
|---|---|---|
| GPQA Diamond | 90.9% | 88.8% |
| SimpleQA Verified | 41% | 33.2% |
| LMArena Expert | 1516 | 1385 |

## Multimodal

- GLM-5.3: —
- Grok 4.3: 31.6 (#104)

| Benchmark | GLM-5.3 | Grok 4.3 |
|---|---|---|
| LMArena Vision | — | 1229 |
| Blueprint-Bench 2 | — | 0% |

## Multilingual

- GLM-5.3: 55.7 (#28)
- Grok 4.3: 50.5 (#120)

| Benchmark | GLM-5.3 | Grok 4.3 |
|---|---|---|
| LMArena Non-English | 1457 | 1385 |
| LMArena Chinese | 1528 | 1422 |
| LMArena French | 1499 | 1412 |
| LMArena German | 1499 | 1395 |
| LMArena Japanese | 1453 | 1379 |
| LMArena Korean | 1472 | 1356 |
| LMArena Russian | 1463 | 1399 |
| LMArena Spanish | 1460 | 1398 |

## Instruction Following

- GLM-5.3: 77.5 (#23)
- Grok 4.3: 72.1 (#140)

| Benchmark | GLM-5.3 | Grok 4.3 |
|---|---|---|
| LMArena Instruction Following | 1477 | 1366 |

## Long Context

- GLM-5.3: 45.4 (#41)
- Grok 4.3: 42.5 (#123)

| Benchmark | GLM-5.3 | Grok 4.3 |
|---|---|---|
| LMArena Longer Query | 1482 | 1393 |

## Writing & Preference

- GLM-5.3: 75.7 (#6)
- Grok 4.3: 58.5 (#118)

| Benchmark | GLM-5.3 | Grok 4.3 |
|---|---|---|
| LMArena Text | 1471 | 1397 |
| LMArena Creative Writing | 1457 | 1380 |
| LMArena Multi-Turn | 1472 | 1406 |
| EQ-Bench Creative Writing | 2075 | — |
| EQ-Bench 4 | — | 1075 |

## FAQ

### Is GLM-5.3 better than Grok 4.3?

GLM-5.3 is the stronger model overall, scoring 54.8 to 43.8 on the Noometry Index.

### Which is cheaper, GLM-5.3 or Grok 4.3?

Grok 4.3 is cheaper. It lists at $1.25 per million input tokens and $2.50 per million output tokens; GLM-5.3 lists at $1.40 and $4.40.

### Is GLM-5.3 or Grok 4.3 better for coding?

GLM-5.3 scores higher on coding benchmarks: 59.5 versus 41.6 in the Noometry coding category.

### Which has the bigger context window?

Both accept 1M tokens.

### How many benchmarks do GLM-5.3 and Grok 4.3 share?

34 benchmarks have published results for both models. GLM-5.3 has 42 scored results on Noometry and Grok 4.3 has 40.