# GLM-4.6 vs Grok 4

> Grok 4 is the stronger model overall, scoring 48.1 to 41.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/glm-4-6-vs-grok-4
- Last updated: 2026-10-11
- Shared benchmarks: 22

## Summary

- They share 22 benchmarks with published results for both. GLM-4.6 scores higher in 3 categories and Grok 4 in 6 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in long context, where Grok 4 leads 63.1 to 43.4.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 47.4% for GLM-4.6 and 73.6% for Grok 4.
- GLM-4.6 has downloadable open weights; the other is API-only.

## Snapshot

| | GLM-4.6 | Grok 4 |
|---|---|---|
| Provider | Z.ai (Zhipu) | xAI |
| Noometry Index | 41.4 | 48.1 |
| Rank | 135 | 56 |
| Context | 205K | — |
| Input $/M | $0.60 | — |
| Output $/M | $2.20 | — |
| Weights | Open | Proprietary |

## Coding

- GLM-4.6: 40.1 (#148)
- Grok 4: 50.3 (#46)

| Benchmark | GLM-4.6 | Grok 4 |
|---|---|---|
| LMArena Coding | 1449 | 1408 |
| SWE-bench Verified (bash only) | 55.4% | — |
| Aider Polyglot | — | 79.6% |
| LMArena WebDev | 1340 | — |
| SciCode | 38.4% | — |
| WeirdML | — | 45.7% |
| ALE-Bench | 340.82 | — |

## Agentic & Tool Use

- GLM-4.6: 32.3 (#66)
- Grok 4: 32.3 (#68)

| Benchmark | GLM-4.6 | Grok 4 |
|---|---|---|
| Terminal-Bench | 24.5% | 27.2% |
| Berkeley Function Calling Leaderboard | 72.4% | 63% |
| GDPval | — | 21.1% |
| Cybench | — | 43% |
| DeepResearch Bench | — | 47.3% |
| BALROG | — | 43.6% |
| LMArena Search | — | 1142 |
| METR Time Horizons | — | 66.6% |

## Reasoning

- GLM-4.6: 23.7 (#172)
- Grok 4: 36.7 (#65)

| Benchmark | GLM-4.6 | Grok 4 |
|---|---|---|
| Kagi LLM Benchmark | 47.4% | 73.6% |
| LMArena Hard Prompts | 1440 | 1409 |
| ARC-AGI-2 | — | 16% |
| SimpleBench | — | 60.5% |
| ARC-AGI-1 | — | 66.7% |
| CritPt | 1.1% | — |
| Chess Puzzles | — | 28% |
| Epoch Capabilities Index | — | 146.44 |
| ForecastBench | — | 60.9 |

## Math

- GLM-4.6: 39.1 (#111)
- Grok 4: 48.4 (#64)

| Benchmark | GLM-4.6 | Grok 4 |
|---|---|---|
| LMArena Math | 1432 | 1422 |
| FrontierMath (Feb 2025 set) | 3.8% | 19.7% |
| FrontierMath Tier 4 (v1) | 2.1% | 2.1% |
| OTIS Mock AIME 2024-2025 | — | 84% |
| Omni-MATH | — | 60.3% |

## Knowledge

- GLM-4.6: 40.2 (#124)
- Grok 4: 53.8 (#55)

| Benchmark | GLM-4.6 | Grok 4 |
|---|---|---|
| LMArena Expert | 1431 | 1415 |
| GPQA Diamond | — | 87% |
| MMLU-Pro | — | 85.1% |
| Confabulations | — | 12.4% |
| Vectara Hallucination Rate | 9.5% | — |
| GPQA (HELM) | — | 72.7% |

## Multimodal

- GLM-4.6: —
- Grok 4: 33.7 (#94)

| Benchmark | GLM-4.6 | Grok 4 |
|---|---|---|
| LMArena Vision | — | 1210 |
| GeoBench | — | 45% |

## Multilingual

- GLM-4.6: 53.5 (#66)
- Grok 4: 51.8 (#103)

| Benchmark | GLM-4.6 | Grok 4 |
|---|---|---|
| LMArena Non-English | 1426 | 1403 |
| LMArena Chinese | 1499 | 1427 |
| LMArena French | 1459 | 1418 |
| LMArena German | 1447 | 1429 |
| LMArena Japanese | 1393 | 1394 |
| LMArena Korean | 1400 | 1377 |
| LMArena Russian | 1419 | 1410 |
| LMArena Spanish | 1436 | 1420 |

## Instruction Following

- GLM-4.6: 74.3 (#98)
- Grok 4: 79.2 (#5)

| Benchmark | GLM-4.6 | Grok 4 |
|---|---|---|
| LMArena Instruction Following | 1410 | 1387 |
| IFEval | — | 94.9% |

## Long Context

- GLM-4.6: 43.4 (#94)
- Grok 4: 63.1 (#4)

| Benchmark | GLM-4.6 | Grok 4 |
|---|---|---|
| LMArena Longer Query | 1422 | 1409 |
| Fiction.LiveBench | — | 94.4% |

## Writing & Preference

- GLM-4.6: 61.1 (#90)
- Grok 4: 58.5 (#116)

| Benchmark | GLM-4.6 | Grok 4 |
|---|---|---|
| LMArena Text | 1440 | 1411 |
| LMArena Creative Writing | 1411 | 1397 |
| LMArena Multi-Turn | 1427 | 1416 |
| Short-Story Creative Writing | — | 76.9% |
| EQ-Bench Creative Writing | 1411 | — |
| WildBench | — | 79.7% |

## FAQ

### Is GLM-4.6 better than Grok 4?

Grok 4 is the stronger model overall, scoring 48.1 to 41.4 on the Noometry Index.

### Is GLM-4.6 or Grok 4 better for coding?

Grok 4 scores higher on coding benchmarks: 50.3 versus 40.1 in the Noometry coding category.

### How many benchmarks do GLM-4.6 and Grok 4 share?

22 benchmarks have published results for both models. GLM-4.6 has 29 scored results on Noometry and Grok 4 has 48.