# GLM-5.2 vs GPT-4.5

> GLM-5.2 is the stronger model overall, scoring 50.9 to 36.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/glm-5-2-vs-gpt-4-5
- Last updated: 2026-10-10
- Shared benchmarks: 24

## Summary

- They share 24 benchmarks with published results for both. GLM-5.2 scores higher in 9 categories and GPT-4.5 in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GLM-5.2 leads 42.1 to 14.1.
- The biggest single-benchmark swing is ARC-AGI-1: 77% for GLM-5.2 and 10.3% for GPT-4.5.
- GLM-5.2 has downloadable open weights; the other is API-only.

## Snapshot

| | GLM-5.2 | GPT-4.5 |
|---|---|---|
| Provider | Z.ai (Zhipu) | OpenAI |
| Noometry Index | 50.9 | 36.9 |
| Rank | 38 | 200 |
| Context | 1M | — |
| Input $/M | $1.40 | — |
| Output $/M | $4.40 | — |
| Weights | Open | Proprietary |

## Coding

- GLM-5.2: 50.9 (#37)
- GPT-4.5: 41.8 (#102)

| Benchmark | GLM-5.2 | GPT-4.5 |
|---|---|---|
| WeirdML | 70.1% | 39.4% |
| LMArena Coding | 1484 | 1397 |
| SWE-bench Verified | 78.7% | — |
| DeepSWE | 43.8% | — |
| FrontierCode | 24.5% | — |
| Aider Polyglot | — | 44.9% |
| LMArena WebDev | 1604 | — |
| SciCode | 50.5% | — |
| LiveBench Coding | — | 75.2% |
| ALE-Bench | 1,047 | — |

## Agentic & Tool Use

- GLM-5.2: 31.6 (#63)
- GPT-4.5: 27.6 (#92)

| Benchmark | GLM-5.2 | GPT-4.5 |
|---|---|---|
| τ²-bench Banking | 37.1% | — |
| Cybench | — | 17.5% |
| PostTrainBench | 31.7% | — |
| GBAEval | 0% | — |
| Vending-Bench 2 | 8,314 | — |

## Reasoning

- GLM-5.2: 42.1 (#46)
- GPT-4.5: 14.1 (#320)

| Benchmark | GLM-5.2 | GPT-4.5 |
|---|---|---|
| ARC-AGI-2 | 22.8% | 0.8% |
| SimpleBench | 58.8% | 34.5% |
| ARC-AGI-1 | 77% | 10.3% |
| LMArena Hard Prompts | 1480 | 1404 |
| Epoch Capabilities Index | 151.76 | 136.74 |
| Kagi LLM Benchmark | 62.6% | — |
| NYT Connections (extended) | 74.3% | — |
| CritPt | 20.9% | — |
| Chess Puzzles | 21% | — |
| EnigmaEval | — | 3.2% |
| EBR-Bench | 9.5% | — |
| LiveBench Reasoning | — | 71.1% |
| Mystery Game Puzzles | 19% | — |
| DTBench | 93.6% | — |
| LiveBench Data Analysis | — | 64.3% |
| LMCA | 45.8% | — |
| Surface Evolver Bench | 55.6% | — |
| ForecastBench | — | 61.7 |
| LiveBench | — | 69% |

## Math

- GLM-5.2: 55.4 (#35)
- GPT-4.5: 31.0 (#200)

| Benchmark | GLM-5.2 | GPT-4.5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 86.4% | 37.8% |
| LMArena Math | 1482 | 1412 |
| FrontierMath (Tiers 1-3) | 59.2% | — |
| FrontierMath Tier 4 | 29.3% | — |
| MathArena Final-Answer Competitions | 67.6% | — |
| ProofBench | 35% | — |
| LiveBench Math | — | 69.3% |
| MATH Level 5 | — | 78.6% |

## Knowledge

- GLM-5.2: 57.1 (#36)
- GPT-4.5: 32.4 (#205)

| Benchmark | GLM-5.2 | GPT-4.5 |
|---|---|---|
| GPQA Diamond | 91.9% | 68.7% |
| LMArena Expert | 1484 | 1394 |
| Humanity's Last Exam | — | 5.4% |
| SimpleQA Verified | 34.2% | — |
| Confabulations | — | 13.6% |

## Multimodal

- GLM-5.2: —
- GPT-4.5: 37.0 (#68)

| Benchmark | GLM-5.2 | GPT-4.5 |
|---|---|---|
| LMArena Vision | — | 1195 |
| VPCT | — | 45% |

## Multilingual

- GLM-5.2: 56.0 (#25)
- GPT-4.5: 52.7 (#78)

| Benchmark | GLM-5.2 | GPT-4.5 |
|---|---|---|
| LMArena Non-English | 1459 | 1413 |
| LMArena Chinese | 1518 | 1421 |
| LMArena French | 1476 | 1417 |
| LMArena German | 1470 | 1457 |
| LMArena Japanese | 1451 | 1416 |
| LMArena Korean | 1445 | 1392 |
| LMArena Russian | 1467 | 1419 |
| LMArena Spanish | 1477 | — |

## Instruction Following

- GLM-5.2: 77.1 (#32)
- GPT-4.5: 72.8 (#129)

| Benchmark | GLM-5.2 | GPT-4.5 |
|---|---|---|
| LMArena Instruction Following | 1465 | 1404 |
| LiveBench Instruction Following | — | 72.3% |

## Long Context

- GLM-5.2: 45.3 (#43)
- GPT-4.5: 40.4 (#150)

| Benchmark | GLM-5.2 | GPT-4.5 |
|---|---|---|
| LMArena Longer Query | 1479 | 1406 |
| Fiction.LiveBench | — | 63.9% |

## Writing & Preference

- GLM-5.2: 70.7 (#20)
- GPT-4.5: 57.1 (#129)

| Benchmark | GLM-5.2 | GPT-4.5 |
|---|---|---|
| LMArena Text | 1470 | 1417 |
| LMArena Creative Writing | 1465 | 1394 |
| EQ-Bench Creative Writing | 1757 | 1258 |
| LMArena Multi-Turn | 1468 | 1444 |
| Short-Story Creative Writing | — | 75.6% |
| EQ-Bench 4 | 1222 | — |
| LiveBench Language | — | 61.5% |

## FAQ

### Is GLM-5.2 better than GPT-4.5?

GLM-5.2 is the stronger model overall, scoring 50.9 to 36.9 on the Noometry Index.

### Is GLM-5.2 or GPT-4.5 better for coding?

GLM-5.2 scores higher on coding benchmarks: 50.9 versus 41.8 in the Noometry coding category.

### How many benchmarks do GLM-5.2 and GPT-4.5 share?

24 benchmarks have published results for both models. GLM-5.2 has 50 scored results on Noometry and GPT-4.5 has 42.