# GLM-5.2 vs GPT-5.6 Terra

> GPT-5.6 Terra is the stronger model overall, scoring 59.2 to 51.1 on the Noometry Index. GLM-5.2 costs 2.1× less per token, which makes it the better buy when GPT-5.6 Terra's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/glm-5-2-vs-gpt-5-6-terra
- Last updated: 2026-10-10
- Shared benchmarks: 45

## Summary

- They share 45 benchmarks with published results for both. GLM-5.2 scores higher in 4 categories and GPT-5.6 Terra in 5 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.6 Terra leads 81.6 to 55.7.
- The biggest single-benchmark swing is ARC-AGI-2: 22.8% for GLM-5.2 and 83.9% for GPT-5.6 Terra.
- GLM-5.2 is cheaper at $1.40 / $4.40 per million input/output tokens, against $2 / $12 for GPT-5.6 Terra.
- GPT-5.6 Terra accepts more context: 1.05M tokens versus 1M.
- GLM-5.2 has downloadable open weights; the other is API-only.

## Snapshot

| | GLM-5.2 | GPT-5.6 Terra |
|---|---|---|
| Provider | Z.ai (Zhipu) | OpenAI |
| Noometry Index | 51.1 | 59.2 |
| Rank | 44 | 17 |
| Context | 1M | 1.05M |
| Input $/M | $1.40 | $2 |
| Output $/M | $4.40 | $12 |
| Weights | Open | Proprietary |

## Coding

- GLM-5.2: 51.3 (#41)
- GPT-5.6 Terra: 57.7 (#19)

| Benchmark | GLM-5.2 | GPT-5.6 Terra |
|---|---|---|
| DeepSWE | 43.8% | 69.6% |
| FrontierCode | 24.5% | 41.3% |
| LMArena WebDev | 1603 | 1522 |
| SciCode | 50.5% | 55% |
| WeirdML | 70.1% | 78.3% |
| LMArena Coding | 1485 | 1484 |
| ALE-Bench | 1,047 | 1,951 |
| SWE-bench Verified | 78.7% | — |
| CursorBench | — | 41.3% |

## Agentic & Tool Use

- GLM-5.2: 32.4 (#63)
- GPT-5.6 Terra: 40.1 (#25)

| Benchmark | GLM-5.2 | GPT-5.6 Terra |
|---|---|---|
| APEX-Agents | 45.2% | 58.2% |
| Vending-Bench 2 | 8,314 | 7,343 |
| τ²-bench Banking | 37.1% | — |
| PostTrainBench | 31.7% | — |
| BALROG | — | 53.2% |
| GBAEval | 0% | — |
| GDP.pdf | — | 24.7% |

## Reasoning

- GLM-5.2: 42.3 (#52)
- GPT-5.6 Terra: 60.7 (#21)

| Benchmark | GLM-5.2 | GPT-5.6 Terra |
|---|---|---|
| ARC-AGI-2 | 22.8% | 83.9% |
| SimpleBench | 58.8% | 48.9% |
| Kagi LLM Benchmark | 62.6% | 51.3% |
| NYT Connections (extended) | 74.3% | 78.4% |
| ARC-AGI-1 | 77% | 96.5% |
| CritPt | 20.9% | 30% |
| Chess Puzzles | 21% | 54% |
| LMArena Hard Prompts | 1480 | 1468 |
| Mystery Game Puzzles | 19% | 35% |
| DTBench | 93.6% | 93.3% |
| LMCA | 45.8% | 55% |
| Surface Evolver Bench | 55.6% | 83.8% |
| Epoch Capabilities Index | 151.78 | 159.62 |
| EBR-Bench | 9.5% | — |

## Math

- GLM-5.2: 55.7 (#43)
- GPT-5.6 Terra: 81.6 (#12)

| Benchmark | GLM-5.2 | GPT-5.6 Terra |
|---|---|---|
| FrontierMath (Tiers 1-3) | 59.2% | 86% |
| FrontierMath Tier 4 | 29.3% | 70.7% |
| OTIS Mock AIME 2024-2025 | 86.4% | 99.7% |
| ProofBench | 35% | 74% |
| LMArena Math | 1482 | 1466 |
| MathArena Final-Answer Competitions | 67.6% | — |

## Knowledge

- GLM-5.2: 57.1 (#40)
- GPT-5.6 Terra: 61.2 (#30)

| Benchmark | GLM-5.2 | GPT-5.6 Terra |
|---|---|---|
| GPQA Diamond | 91.9% | 93.3% |
| SimpleQA Verified | 34.2% | 43.2% |
| LMArena Expert | 1486 | 1492 |

## Multimodal

- GLM-5.2: —
- GPT-5.6 Terra: 47.3 (#11)

| Benchmark | GLM-5.2 | GPT-5.6 Terra |
|---|---|---|
| LMArena Vision | — | 1271 |
| Blueprint-Bench 2 | — | 30.8% |
| Furniture Assembly | — | 54.2% |
| LMArena Document | — | 1472 |

## Multilingual

- GLM-5.2: 55.8 (#26)
- GPT-5.6 Terra: 54.4 (#44)

| Benchmark | GLM-5.2 | GPT-5.6 Terra |
|---|---|---|
| LMArena Non-English | 1459 | 1439 |
| LMArena Chinese | 1519 | 1513 |
| LMArena French | 1479 | 1471 |
| LMArena German | 1468 | 1460 |
| LMArena Japanese | 1451 | 1457 |
| LMArena Korean | 1445 | 1425 |
| LMArena Russian | 1466 | 1450 |
| LMArena Spanish | 1477 | 1448 |

## Instruction Following

- GLM-5.2: 76.9 (#34)
- GPT-5.6 Terra: 76.4 (#40)

| Benchmark | GLM-5.2 | GPT-5.6 Terra |
|---|---|---|
| LMArena Instruction Following | 1465 | 1454 |

## Long Context

- GLM-5.2: 45.3 (#43)
- GPT-5.6 Terra: 44.4 (#68)

| Benchmark | GLM-5.2 | GPT-5.6 Terra |
|---|---|---|
| LMArena Longer Query | 1479 | 1451 |

## Writing & Preference

- GLM-5.2: 70.4 (#21)
- GPT-5.6 Terra: 70.2 (#23)

| Benchmark | GLM-5.2 | GPT-5.6 Terra |
|---|---|---|
| LMArena Text | 1470 | 1447 |
| LMArena Creative Writing | 1462 | 1410 |
| EQ-Bench Creative Writing | 1757 | 1855 |
| EQ-Bench 4 | 1222 | 1234 |
| LMArena Multi-Turn | 1469 | 1449 |

## FAQ

### Is GLM-5.2 better than GPT-5.6 Terra?

GPT-5.6 Terra is the stronger model overall, scoring 59.2 to 51.1 on the Noometry Index. GLM-5.2 costs 2.1× less per token, which makes it the better buy when GPT-5.6 Terra's lead doesn't matter for your workload.

### Which is cheaper, GLM-5.2 or GPT-5.6 Terra?

GLM-5.2 is cheaper. It lists at $1.40 per million input tokens and $4.40 per million output tokens; GPT-5.6 Terra lists at $2 and $12.

### Is GLM-5.2 or GPT-5.6 Terra better for coding?

GPT-5.6 Terra scores higher on coding benchmarks: 57.7 versus 51.3 in the Noometry coding category.

### Which has the bigger context window?

GPT-5.6 Terra does, with 1.05M tokens against 1M.

### How many benchmarks do GLM-5.2 and GPT-5.6 Terra share?

45 benchmarks have published results for both models. GLM-5.2 has 51 scored results on Noometry and GPT-5.6 Terra has 52.