# GLM-4.6 vs GLM-5.2

> GLM-5.2 is the stronger model overall, scoring 51.1 to 41.4 on the Noometry Index. GLM-4.6 costs 2.1× less per token, which makes it the better buy when GLM-5.2's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/glm-4-6-vs-glm-5-2
- Last updated: 2026-10-11
- Shared benchmarks: 23

## Summary

- They share 23 benchmarks with published results for both. GLM-4.6 scores higher in 0 categories and GLM-5.2 in 9 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GLM-5.2 leads 42.3 to 23.7.
- The biggest single-benchmark swing is CritPt: 1.1% for GLM-4.6 and 20.9% for GLM-5.2.
- GLM-4.6 is cheaper at $0.60 / $2.20 per million input/output tokens, against $1.40 / $4.40 for GLM-5.2.
- GLM-5.2 accepts more context: 1M tokens versus 205K.

## Snapshot

| | GLM-4.6 | GLM-5.2 |
|---|---|---|
| Provider | Z.ai (Zhipu) | Z.ai (Zhipu) |
| Noometry Index | 41.4 | 51.1 |
| Rank | 135 | 44 |
| Context | 205K | 1M |
| Input $/M | $0.60 | $1.40 |
| Output $/M | $2.20 | $4.40 |
| Weights | Open | Open |

## Coding

- GLM-4.6: 40.1 (#148)
- GLM-5.2: 51.3 (#41)

| Benchmark | GLM-4.6 | GLM-5.2 |
|---|---|---|
| LMArena WebDev | 1340 | 1603 |
| SciCode | 38.4% | 50.5% |
| LMArena Coding | 1449 | 1485 |
| ALE-Bench | 340.82 | 1,047 |
| SWE-bench Verified | — | 78.7% |
| DeepSWE | — | 43.8% |
| FrontierCode | — | 24.5% |
| SWE-bench Verified (bash only) | 55.4% | — |
| WeirdML | — | 70.1% |

## Agentic & Tool Use

- GLM-4.6: 32.3 (#66)
- GLM-5.2: 32.4 (#63)

| Benchmark | GLM-4.6 | GLM-5.2 |
|---|---|---|
| Terminal-Bench | 24.5% | — |
| APEX-Agents | — | 45.2% |
| Berkeley Function Calling Leaderboard | 72.4% | — |
| τ²-bench Banking | — | 37.1% |
| PostTrainBench | — | 31.7% |
| GBAEval | — | 0% |
| Vending-Bench 2 | — | 8,314 |

## Reasoning

- GLM-4.6: 23.7 (#172)
- GLM-5.2: 42.3 (#52)

| Benchmark | GLM-4.6 | GLM-5.2 |
|---|---|---|
| Kagi LLM Benchmark | 47.4% | 62.6% |
| CritPt | 1.1% | 20.9% |
| LMArena Hard Prompts | 1440 | 1480 |
| ARC-AGI-2 | — | 22.8% |
| SimpleBench | — | 58.8% |
| NYT Connections (extended) | — | 74.3% |
| ARC-AGI-1 | — | 77% |
| Chess Puzzles | — | 21% |
| EBR-Bench | — | 9.5% |
| Mystery Game Puzzles | — | 19% |
| DTBench | — | 93.6% |
| LMCA | — | 45.8% |
| Surface Evolver Bench | — | 55.6% |
| Epoch Capabilities Index | — | 151.78 |

## Math

- GLM-4.6: 39.1 (#111)
- GLM-5.2: 55.7 (#43)

| Benchmark | GLM-4.6 | GLM-5.2 |
|---|---|---|
| LMArena Math | 1432 | 1482 |
| FrontierMath (Tiers 1-3) | — | 59.2% |
| FrontierMath Tier 4 | — | 29.3% |
| MathArena Final-Answer Competitions | — | 67.6% |
| OTIS Mock AIME 2024-2025 | — | 86.4% |
| ProofBench | — | 35% |
| FrontierMath (Feb 2025 set) | 3.8% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |

## Knowledge

- GLM-4.6: 40.2 (#124)
- GLM-5.2: 57.1 (#40)

| Benchmark | GLM-4.6 | GLM-5.2 |
|---|---|---|
| LMArena Expert | 1431 | 1486 |
| GPQA Diamond | — | 91.9% |
| SimpleQA Verified | — | 34.2% |
| Vectara Hallucination Rate | 9.5% | — |

## Multilingual

- GLM-4.6: 53.5 (#66)
- GLM-5.2: 55.8 (#26)

| Benchmark | GLM-4.6 | GLM-5.2 |
|---|---|---|
| LMArena Non-English | 1426 | 1459 |
| LMArena Chinese | 1499 | 1519 |
| LMArena French | 1459 | 1479 |
| LMArena German | 1447 | 1468 |
| LMArena Japanese | 1393 | 1451 |
| LMArena Korean | 1400 | 1445 |
| LMArena Russian | 1419 | 1466 |
| LMArena Spanish | 1436 | 1477 |

## Instruction Following

- GLM-4.6: 74.3 (#98)
- GLM-5.2: 76.9 (#34)

| Benchmark | GLM-4.6 | GLM-5.2 |
|---|---|---|
| LMArena Instruction Following | 1410 | 1465 |

## Long Context

- GLM-4.6: 43.4 (#94)
- GLM-5.2: 45.3 (#43)

| Benchmark | GLM-4.6 | GLM-5.2 |
|---|---|---|
| LMArena Longer Query | 1422 | 1479 |

## Writing & Preference

- GLM-4.6: 61.1 (#90)
- GLM-5.2: 70.4 (#21)

| Benchmark | GLM-4.6 | GLM-5.2 |
|---|---|---|
| LMArena Text | 1440 | 1470 |
| LMArena Creative Writing | 1411 | 1462 |
| EQ-Bench Creative Writing | 1411 | 1757 |
| LMArena Multi-Turn | 1427 | 1469 |
| EQ-Bench 4 | — | 1222 |

## FAQ

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

GLM-5.2 is the stronger model overall, scoring 51.1 to 41.4 on the Noometry Index. GLM-4.6 costs 2.1× less per token, which makes it the better buy when GLM-5.2's lead doesn't matter for your workload.

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

GLM-4.6 is cheaper. It lists at $0.60 per million input tokens and $2.20 per million output tokens; GLM-5.2 lists at $1.40 and $4.40.

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

GLM-5.2 scores higher on coding benchmarks: 51.3 versus 40.1 in the Noometry coding category.

### Which has the bigger context window?

GLM-5.2 does, with 1M tokens against 205K.

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

23 benchmarks have published results for both models. GLM-4.6 has 29 scored results on Noometry and GLM-5.2 has 51.