# GLM-5.2 vs Grok 4.6

> Grok 4.6 is the stronger model overall, scoring 56.8 to 50.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/glm-5-2-vs-grok-4-6
- Last updated: 2026-10-10
- Shared benchmarks: 41

## Summary

- They share 41 benchmarks with published results for both. GLM-5.2 scores higher in 4 categories and Grok 4.6 in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.6 leads 61.3 to 42.1.
- The biggest single-benchmark swing is ARC-AGI-2: 22.8% for GLM-5.2 and 67.1% for Grok 4.6.
- GLM-5.2 is cheaper at $1.40 / $4.40 per million input/output tokens, against $2 / $6 for Grok 4.6.
- GLM-5.2 accepts more context: 1M tokens versus 500K.
- GLM-5.2 has downloadable open weights; the other is API-only.

## Snapshot

| | GLM-5.2 | Grok 4.6 |
|---|---|---|
| Provider | Z.ai (Zhipu) | xAI |
| Noometry Index | 50.9 | 56.8 |
| Rank | 38 | 19 |
| Context | 1M | 500K |
| Input $/M | $1.40 | $2 |
| Output $/M | $4.40 | $6 |
| Weights | Open | Proprietary |

## Coding

- GLM-5.2: 50.9 (#37)
- Grok 4.6: 58.0 (#14)

| Benchmark | GLM-5.2 | Grok 4.6 |
|---|---|---|
| DeepSWE | 43.8% | 67.5% |
| FrontierCode | 24.5% | 48% |
| LMArena WebDev | 1604 | 1620 |
| SciCode | 50.5% | 56.5% |
| WeirdML | 70.1% | 67.3% |
| LMArena Coding | 1484 | 1464 |
| ALE-Bench | 1,047 | 1,508 |
| SWE-bench Verified | 78.7% | — |
| CursorBench | — | 41.4% |
| FrontierSWE | — | 25.3% |

## Agentic & Tool Use

- GLM-5.2: 31.6 (#63)
- Grok 4.6: 39.4 (#25)

| Benchmark | GLM-5.2 | Grok 4.6 |
|---|---|---|
| Vending-Bench 2 | 8,314 | 9,047 |
| APEX-Agents | — | 65.3% |
| τ²-bench Banking | 37.1% | — |
| PostTrainBench | 31.7% | — |
| GBAEval | 0% | — |
| GDP.pdf | — | 17.2% |

## Reasoning

- GLM-5.2: 42.1 (#46)
- Grok 4.6: 61.3 (#18)

| Benchmark | GLM-5.2 | Grok 4.6 |
|---|---|---|
| ARC-AGI-2 | 22.8% | 67.1% |
| SimpleBench | 58.8% | 75.9% |
| NYT Connections (extended) | 74.3% | 80% |
| ARC-AGI-1 | 77% | 87.5% |
| CritPt | 20.9% | 19.7% |
| Chess Puzzles | 21% | 40% |
| EBR-Bench | 9.5% | 30.5% |
| LMArena Hard Prompts | 1480 | 1445 |
| Mystery Game Puzzles | 19% | 34% |
| DTBench | 93.6% | 97.3% |
| LMCA | 45.8% | 48.5% |
| Epoch Capabilities Index | 151.76 | 156.48 |
| Kagi LLM Benchmark | 62.6% | — |
| Surface Evolver Bench | 55.6% | — |

## Math

- GLM-5.2: 55.4 (#35)
- Grok 4.6: 66.8 (#19)

| Benchmark | GLM-5.2 | Grok 4.6 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 59.2% | 66% |
| FrontierMath Tier 4 | 29.3% | 31.7% |
| OTIS Mock AIME 2024-2025 | 86.4% | 99.2% |
| ProofBench | 35% | 51% |
| LMArena Math | 1482 | 1416 |
| MathArena Final-Answer Competitions | 67.6% | — |

## Knowledge

- GLM-5.2: 57.1 (#36)
- Grok 4.6: 63.3 (#18)

| Benchmark | GLM-5.2 | Grok 4.6 |
|---|---|---|
| GPQA Diamond | 91.9% | 94% |
| SimpleQA Verified | 34.2% | 49.3% |
| LMArena Expert | 1484 | 1463 |

## Multimodal

- GLM-5.2: —
- Grok 4.6: 43.9 (#18)

| Benchmark | GLM-5.2 | Grok 4.6 |
|---|---|---|
| LMArena Vision | — | 1265 |
| Blueprint-Bench 2 | — | 33.2% |
| Furniture Assembly | — | 40% |
| LMArena Document | — | 1452 |

## Multilingual

- GLM-5.2: 56.0 (#25)
- Grok 4.6: 53.3 (#67)

| Benchmark | GLM-5.2 | Grok 4.6 |
|---|---|---|
| LMArena Non-English | 1459 | 1422 |
| LMArena Chinese | 1518 | 1479 |
| LMArena French | 1476 | 1458 |
| LMArena German | 1470 | 1425 |
| LMArena Japanese | 1451 | 1383 |
| LMArena Korean | 1445 | 1395 |
| LMArena Russian | 1467 | 1427 |
| LMArena Spanish | 1477 | 1408 |

## Instruction Following

- GLM-5.2: 77.1 (#32)
- Grok 4.6: 75.6 (#59)

| Benchmark | GLM-5.2 | Grok 4.6 |
|---|---|---|
| LMArena Instruction Following | 1465 | 1431 |

## Long Context

- GLM-5.2: 45.3 (#43)
- Grok 4.6: 44.4 (#62)

| Benchmark | GLM-5.2 | Grok 4.6 |
|---|---|---|
| LMArena Longer Query | 1479 | 1453 |

## Writing & Preference

- GLM-5.2: 70.7 (#20)
- Grok 4.6: 62.5 (#77)

| Benchmark | GLM-5.2 | Grok 4.6 |
|---|---|---|
| LMArena Text | 1470 | 1427 |
| LMArena Creative Writing | 1465 | 1430 |
| LMArena Multi-Turn | 1468 | 1422 |
| EQ-Bench Creative Writing | 1757 | — |
| EQ-Bench 4 | 1222 | — |

## FAQ

### Is GLM-5.2 better than Grok 4.6?

Grok 4.6 is the stronger model overall, scoring 56.8 to 50.9 on the Noometry Index.

### Which is cheaper, GLM-5.2 or Grok 4.6?

GLM-5.2 is cheaper. It lists at $1.40 per million input tokens and $4.40 per million output tokens; Grok 4.6 lists at $2 and $6.

### Is GLM-5.2 or Grok 4.6 better for coding?

Grok 4.6 scores higher on coding benchmarks: 58.0 versus 50.9 in the Noometry coding category.

### Which has the bigger context window?

GLM-5.2 does, with 1M tokens against 500K.

### How many benchmarks do GLM-5.2 and Grok 4.6 share?

41 benchmarks have published results for both models. GLM-5.2 has 50 scored results on Noometry and Grok 4.6 has 49.