# GLM-5.3-Flash vs Grok-3 mini

> GLM-5.3-Flash is the stronger model overall, scoring 51.8 to 41.2 on the Noometry Index.

- Canonical page: https://noometry.com/compare/glm-5-3-flash-vs-grok-3-mini
- Last updated: 2026-10-10
- Shared benchmarks: 22

## Summary

- They share 22 benchmarks with published results for both. GLM-5.3-Flash scores higher in 7 categories and Grok-3 mini in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GLM-5.3-Flash leads 48.0 to 13.6.
- The biggest single-benchmark swing is ARC-AGI-1: 91% for GLM-5.3-Flash and 16.5% for Grok-3 mini.
- GLM-5.3-Flash has downloadable open weights; the other is API-only.

## Snapshot

| | GLM-5.3-Flash | Grok-3 mini |
|---|---|---|
| Provider | Z.ai (Zhipu) | xAI |
| Noometry Index | 51.8 | 41.2 |
| Rank | 41 | 141 |
| Context | 1M | — |
| Input $/M | $0.15 | — |
| Output $/M | $0.50 | — |
| Weights | Open | Proprietary |

## Coding

- GLM-5.3-Flash: 53.1 (#31)
- Grok-3 mini: 40.8 (#131)

| Benchmark | GLM-5.3-Flash | Grok-3 mini |
|---|---|---|
| LMArena Coding | 1508 | 1379 |
| DeepSWE | 63.4% | — |
| FrontierCode | 31.8% | — |
| Aider Polyglot | — | 49.3% |
| CursorBench | 36.8% | — |
| LMArena WebDev | 1609 | — |
| FrontierSWE | 18.1% | — |
| SciCode | 51.6% | — |
| WeirdML | — | 42.6% |
| ALE-Bench | 303.55 | — |

## Agentic & Tool Use

- GLM-5.3-Flash: 34.2 (#47)
- Grok-3 mini: —

| Benchmark | GLM-5.3-Flash | Grok-3 mini |
|---|---|---|
| APEX-Agents | 52.8% | — |
| GDP.pdf | 14% | — |

## Reasoning

- GLM-5.3-Flash: 48.0 (#42)
- Grok-3 mini: 13.6 (#334)

| Benchmark | GLM-5.3-Flash | Grok-3 mini |
|---|---|---|
| ARC-AGI-2 | 65.8% | 0.4% |
| ARC-AGI-1 | 91% | 16.5% |
| LMArena Hard Prompts | 1491 | 1375 |
| Epoch Capabilities Index | 151.88 | 140.35 |
| Kagi LLM Benchmark | — | 61.3% |
| CritPt | 15.4% | — |
| Chess Puzzles | 14% | — |
| Mystery Game Puzzles | 8% | — |
| Surface Evolver Bench | 52.5% | — |
| Bench to the Future 3 | 0.15 | — |

## Math

- GLM-5.3-Flash: 53.3 (#47)
- Grok-3 mini: 42.1 (#85)

| Benchmark | GLM-5.3-Flash | Grok-3 mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 93.9% | 77.8% |
| LMArena Math | 1500 | 1386 |
| FrontierMath (Tiers 1-3) | 55.8% | — |
| FrontierMath Tier 4 | 17.1% | — |
| ProofBench | 21% | — |
| Omni-MATH | — | 31.8% |
| MATH Level 5 | — | 90.9% |
| FrontierMath (Feb 2025 set) | — | 5.9% |

## Knowledge

- GLM-5.3-Flash: 58.4 (#36)
- Grok-3 mini: 46.4 (#81)

| Benchmark | GLM-5.3-Flash | Grok-3 mini |
|---|---|---|
| GPQA Diamond | 90.2% | 76.3% |
| LMArena Expert | 1513 | 1395 |
| MMLU-Pro | — | 79.9% |
| Confabulations | — | 10.8% |
| GPQA (HELM) | — | 67.5% |

## Multimodal

- GLM-5.3-Flash: 42.8 (#27)
- Grok-3 mini: —

| Benchmark | GLM-5.3-Flash | Grok-3 mini |
|---|---|---|
| LMArena Vision | 1296 | — |

## Multilingual

- GLM-5.3-Flash: 56.0 (#25)
- Grok-3 mini: 48.1 (#145)

| Benchmark | GLM-5.3-Flash | Grok-3 mini |
|---|---|---|
| LMArena Non-English | 1462 | 1352 |
| LMArena Chinese | 1527 | 1387 |
| LMArena French | 1496 | 1357 |
| LMArena German | 1470 | 1349 |
| LMArena Japanese | 1429 | 1342 |
| LMArena Korean | 1446 | 1335 |
| LMArena Russian | 1469 | 1353 |
| LMArena Spanish | 1471 | 1381 |

## Instruction Following

- GLM-5.3-Flash: 77.5 (#20)
- Grok-3 mini: 78.5 (#9)

| Benchmark | GLM-5.3-Flash | Grok-3 mini |
|---|---|---|
| LMArena Instruction Following | 1478 | 1357 |
| IFEval | — | 95.1% |

## Long Context

- GLM-5.3-Flash: 45.4 (#39)
- Grok-3 mini: 41.0 (#147)

| Benchmark | GLM-5.3-Flash | Grok-3 mini |
|---|---|---|
| LMArena Longer Query | 1482 | 1372 |
| Fiction.LiveBench | — | 66.7% |

## Writing & Preference

- GLM-5.3-Flash: 65.3 (#50)
- Grok-3 mini: 52.5 (#169)

| Benchmark | GLM-5.3-Flash | Grok-3 mini |
|---|---|---|
| LMArena Text | 1471 | 1370 |
| LMArena Creative Writing | 1442 | 1342 |
| LMArena Multi-Turn | 1467 | 1355 |
| Short-Story Creative Writing | — | 73.5% |
| WildBench | — | 65.1% |

## FAQ

### Is GLM-5.3-Flash better than Grok-3 mini?

GLM-5.3-Flash is the stronger model overall, scoring 51.8 to 41.2 on the Noometry Index.

### Is GLM-5.3-Flash or Grok-3 mini better for coding?

GLM-5.3-Flash scores higher on coding benchmarks: 53.1 versus 40.8 in the Noometry coding category.

### How many benchmarks do GLM-5.3-Flash and Grok-3 mini share?

22 benchmarks have published results for both models. GLM-5.3-Flash has 40 scored results on Noometry and Grok-3 mini has 35.