# GLM-4.6 vs Qwen2.5 72B Instruct

> GLM-4.6 is the stronger model overall, scoring 40.9 to 31.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/glm-4-6-vs-qwen2-5-72b-instruct
- Last updated: 2026-10-10
- Shared benchmarks: 17

## Summary

- They share 17 benchmarks with published results for both. GLM-4.6 scores higher in 9 categories and Qwen2.5 72B Instruct in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GLM-4.6 leads 36.4 to 18.1.
- GLM-4.6 is cheaper at $0.60 / $2.20 per million input/output tokens, against $1.40 / $5.60 for Qwen2.5 72B Instruct.
- GLM-4.6 accepts more context: 205K tokens versus 131K.

## Snapshot

| | GLM-4.6 | Qwen2.5 72B Instruct |
|---|---|---|
| Provider | Z.ai (Zhipu) | Alibaba (Qwen) |
| Noometry Index | 40.9 | 31.5 |
| Rank | 132 | 259 |
| Context | 205K | 131K |
| Input $/M | $0.60 | $1.40 |
| Output $/M | $2.20 | $5.60 |
| Weights | Open | Open |

## Coding

- GLM-4.6: 39.9 (#137)
- Qwen2.5 72B Instruct: 32.8 (#256)

| Benchmark | GLM-4.6 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Coding | 1449 | 1293 |
| SWE-bench Verified (bash only) | 55.4% | — |
| LMArena WebDev | 1339 | — |
| SciCode | 38.4% | — |
| WeirdML | — | 16% |
| BigCodeBench Instruct | — | 45.8% |
| BigCodeBench Complete | — | 55.9% |
| ALE-Bench | 340.82 | — |

## Agentic & Tool Use

- GLM-4.6: 32.1 (#60)
- Qwen2.5 72B Instruct: 22.0 (#124)

| Benchmark | GLM-4.6 | Qwen2.5 72B Instruct |
|---|---|---|
| Terminal-Bench | 24.5% | — |
| Berkeley Function Calling Leaderboard | 72.4% | — |
| TheAgentCompany | — | 5.7% |
| BALROG | — | 16.2% |
| METR Time Horizons | — | 35.8% |

## Reasoning

- GLM-4.6: 23.1 (#168)
- Qwen2.5 72B Instruct: 21.4 (#197)

| Benchmark | GLM-4.6 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1440 | 1271 |
| Kagi LLM Benchmark | 47.4% | — |
| CritPt | 1.1% | — |
| DTBench | — | 62.9% |
| LMCA | — | 13.4% |
| BIG-Bench Hard | — | 79.8% |
| Epoch Capabilities Index | — | 129.01 |
| ForecastBench | — | 57.5 |
| HellaSwag | — | 84.8% |
| PIQA | — | 82.6% |
| WinoGrande | — | 82.3% |

## Math

- GLM-4.6: 36.4 (#115)
- Qwen2.5 72B Instruct: 18.1 (#281)

| Benchmark | GLM-4.6 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Math | 1431 | 1283 |
| OTIS Mock AIME 2024-2025 | — | 8.1% |
| Omni-MATH | — | 33% |
| MATH Level 5 | — | 63.2% |
| FrontierMath (Feb 2025 set) | 3.8% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |

## Knowledge

- GLM-4.6: 39.6 (#118)
- Qwen2.5 72B Instruct: 26.9 (#247)

| Benchmark | GLM-4.6 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Expert | 1432 | 1245 |
| GPQA Diamond | — | 49.1% |
| MMLU-Pro | — | 63.1% |
| Confabulations | — | 19.1% |
| Vectara Hallucination Rate | 9.5% | — |
| GPQA (HELM) | — | 42.6% |
| ARC (AI2) Challenge | — | 94.5% |
| MMLU | — | 85.3% |
| TriviaQA | — | 71.9% |

## Multilingual

- GLM-4.6: 53.6 (#62)
- Qwen2.5 72B Instruct: 41.2 (#208)

| Benchmark | GLM-4.6 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1426 | 1252 |
| LMArena Chinese | 1500 | 1272 |
| LMArena French | 1459 | 1279 |
| LMArena German | 1447 | 1234 |
| LMArena Japanese | 1394 | 1180 |
| LMArena Korean | 1401 | 1188 |
| LMArena Russian | 1418 | 1264 |
| LMArena Spanish | 1436 | 1255 |

## Instruction Following

- GLM-4.6: 74.5 (#93)
- Qwen2.5 72B Instruct: 65.7 (#216)

| Benchmark | GLM-4.6 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Instruction Following | 1410 | 1254 |
| IFEval | — | 80.6% |

## Long Context

- GLM-4.6: 43.4 (#91)
- Qwen2.5 72B Instruct: 38.9 (#183)

| Benchmark | GLM-4.6 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1423 | 1282 |

## Writing & Preference

- GLM-4.6: 61.3 (#86)
- Qwen2.5 72B Instruct: 46.9 (#210)

| Benchmark | GLM-4.6 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1440 | 1269 |
| LMArena Creative Writing | 1411 | 1221 |
| LMArena Multi-Turn | 1427 | 1272 |
| EQ-Bench Creative Writing | 1411 | — |
| WildBench | — | 80.2% |

## FAQ

### Is GLM-4.6 better than Qwen2.5 72B Instruct?

GLM-4.6 is the stronger model overall, scoring 40.9 to 31.5 on the Noometry Index.

### Which is cheaper, GLM-4.6 or Qwen2.5 72B Instruct?

GLM-4.6 is cheaper. It lists at $0.60 per million input tokens and $2.20 per million output tokens; Qwen2.5 72B Instruct lists at $1.40 and $5.60.

### Is GLM-4.6 or Qwen2.5 72B Instruct better for coding?

GLM-4.6 scores higher on coding benchmarks: 39.9 versus 32.8 in the Noometry coding category.

### Which has the bigger context window?

GLM-4.6 does, with 205K tokens against 131K.

### How many benchmarks do GLM-4.6 and Qwen2.5 72B Instruct share?

17 benchmarks have published results for both models. GLM-4.6 has 29 scored results on Noometry and Qwen2.5 72B Instruct has 43.