# GLM-5 vs GPT-4

> GLM-5 is the stronger model overall, scoring 46.1 to 29.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/glm-5-vs-gpt-4
- Last updated: 2026-10-11
- Shared benchmarks: 24

## Summary

- They share 24 benchmarks with published results for both. GLM-5 scores higher in 8 categories and GPT-4 in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GLM-5 leads 46.4 to 10.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 80% for GLM-5 and 1.1% for GPT-4.
- GLM-5 is cheaper at $1 / $3.20 per million input/output tokens, against $30 / $60 for GPT-4.
- GLM-5 accepts more context: 205K tokens versus 8K.
- GLM-5 has downloadable open weights; the other is API-only.

## Snapshot

| | GLM-5 | GPT-4 |
|---|---|---|
| Provider | Z.ai (Zhipu) | OpenAI |
| Noometry Index | 46.1 | 29.1 |
| Rank | 66 | 316 |
| Context | 205K | 8K |
| Input $/M | $1 | $30 |
| Output $/M | $3.20 | $60 |
| Weights | Open | Proprietary |

## Coding

- GLM-5: 49.0 (#52)
- GPT-4: 31.6 (#283)

| Benchmark | GLM-5 | GPT-4 |
|---|---|---|
| WeirdML | 48.2% | 12.4% |
| LMArena Coding | 1461 | 1254 |
| SWE-bench Verified | 72.1% | — |
| SWE-bench Verified (bash only) | 72.8% | — |
| LMArena WebDev | 1434 | — |
| SWE-bench Multilingual | 69.7% | — |
| BigCodeBench Instruct | — | 46% |
| BigCodeBench Complete | — | 57.2% |
| ALE-Bench | 765.62 | — |
| HumanEval+ | — | 79.3% |

## Agentic & Tool Use

- GLM-5: 31.1 (#71)
- GPT-4: —

| Benchmark | GLM-5 | GPT-4 |
|---|---|---|
| Terminal-Bench | 52.4% | — |
| τ²-bench Airline | 82.5% | — |
| τ²-bench Banking | 9.8% | — |
| τ²-bench Retail | 73.7% | — |
| τ²-bench Telecom | 86.8% | — |
| METR Time Horizons | — | 36.1% |
| Vending-Bench 2 | 4,432 | — |

## Reasoning

- GLM-5: 27.6 (#116)
- GPT-4: 17.8 (#289)

| Benchmark | GLM-5 | GPT-4 |
|---|---|---|
| Chess Puzzles | 10% | 4% |
| LMArena Hard Prompts | 1452 | 1241 |
| Epoch Capabilities Index | 145.83 | 125.89 |
| ForecastBench | 61 | 57.8 |
| ARC-AGI-2 | 4.9% | — |
| SimpleBench | 53.2% | — |
| Kagi LLM Benchmark | 75% | — |
| NYT Connections (extended) | 74.8% | — |
| ARC-AGI-1 | 44.7% | — |
| Mystery Game Puzzles | — | 12% |
| DTBench | — | 62.7% |
| LMCA | — | 17.1% |
| BIG-Bench Hard | — | 75.1% |
| HellaSwag | — | 95.3% |
| WinoGrande | — | 87.5% |

## Math

- GLM-5: 46.4 (#71)
- GPT-4: 10.8 (#309)

| Benchmark | GLM-5 | GPT-4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 80% | 1.1% |
| LMArena Math | 1440 | 1269 |
| MathArena Final-Answer Competitions | 65.7% | — |
| MATH Level 5 | — | 23% |
| FrontierMath (Feb 2025 set) | 16.4% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
| GSM8K | — | 92% |

## Knowledge

- GLM-5: 52.3 (#64)
- GPT-4: 18.4 (#282)

| Benchmark | GLM-5 | GPT-4 |
|---|---|---|
| GPQA Diamond | 87.8% | 35.7% |
| LMArena Expert | 1454 | 1211 |
| Vectara Hallucination Rate | 10.1% | — |
| MMLU | — | 86.4% |
| TriviaQA | — | 84.8% |

## Multilingual

- GLM-5: 53.7 (#58)
- GPT-4: 40.6 (#215)

| Benchmark | GLM-5 | GPT-4 |
|---|---|---|
| LMArena Non-English | 1430 | 1246 |
| LMArena Chinese | 1511 | 1242 |
| LMArena French | 1455 | 1283 |
| LMArena German | 1445 | 1251 |
| LMArena Japanese | 1416 | 1209 |
| LMArena Korean | 1423 | 1184 |
| LMArena Russian | 1436 | 1251 |
| LMArena Spanish | 1454 | 1261 |

## Instruction Following

- GLM-5: 75.2 (#67)
- GPT-4: 65.3 (#222)

| Benchmark | GLM-5 | GPT-4 |
|---|---|---|
| LMArena Instruction Following | 1428 | 1241 |

## Long Context

- GLM-5: 44.7 (#60)
- GPT-4: 37.7 (#212)

| Benchmark | GLM-5 | GPT-4 |
|---|---|---|
| LMArena Longer Query | 1446 | 1244 |
| CL-bench | 18.7% | — |

## Writing & Preference

- GLM-5: 66.0 (#38)
- GPT-4: 34.9 (#268)

| Benchmark | GLM-5 | GPT-4 |
|---|---|---|
| LMArena Text | 1446 | 1263 |
| LMArena Creative Writing | 1439 | 1244 |
| EQ-Bench Creative Writing | 1601 | 752 |
| LMArena Multi-Turn | 1456 | 1257 |

## FAQ

### Is GLM-5 better than GPT-4?

GLM-5 is the stronger model overall, scoring 46.1 to 29.1 on the Noometry Index.

### Which is cheaper, GLM-5 or GPT-4?

GLM-5 is cheaper. It lists at $1 per million input tokens and $3.20 per million output tokens; GPT-4 lists at $30 and $60.

### Is GLM-5 or GPT-4 better for coding?

GLM-5 scores higher on coding benchmarks: 49.0 versus 31.6 in the Noometry coding category.

### Which has the bigger context window?

GLM-5 does, with 205K tokens against 8K.

### How many benchmarks do GLM-5 and GPT-4 share?

24 benchmarks have published results for both models. GLM-5 has 45 scored results on Noometry and GPT-4 has 38.