# GLM-4.6 vs QwQ-32B

> GLM-4.6 is the stronger model overall, scoring 40.9 to 39.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/glm-4-6-vs-qwq-32b
- Last updated: 2026-10-10
- Shared benchmarks: 18

## Summary

- They share 18 benchmarks with published results for both. GLM-4.6 scores higher in 7 categories and QwQ-32B in 1 category; 6 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GLM-4.6 leads 61.3 to 50.9.

## Snapshot

| | GLM-4.6 | QwQ-32B |
|---|---|---|
| Provider | Z.ai (Zhipu) | Alibaba (Qwen) |
| Noometry Index | 40.9 | 39.4 |
| Rank | 132 | 153 |
| Context | 205K | — |
| Input $/M | $0.60 | — |
| Output $/M | $2.20 | — |
| Weights | Open | Open |

## Coding

- GLM-4.6: 39.9 (#137)
- QwQ-32B: 34.9 (#220)

| Benchmark | GLM-4.6 | QwQ-32B |
|---|---|---|
| LMArena Coding | 1449 | 1333 |
| SWE-bench Verified (bash only) | 55.4% | — |
| Aider Polyglot | — | 20.9% |
| LMArena WebDev | 1339 | — |
| SciCode | 38.4% | — |
| BigCodeBench Instruct | — | 44.6% |
| LiveBench Coding | — | 72.2% |
| BigCodeBench Complete | — | 54.4% |
| ALE-Bench | 340.82 | — |

## Agentic & Tool Use

- GLM-4.6: 32.1 (#60)
- QwQ-32B: —

| Benchmark | GLM-4.6 | QwQ-32B |
|---|---|---|
| Terminal-Bench | 24.5% | — |
| Berkeley Function Calling Leaderboard | 72.4% | — |

## Reasoning

- GLM-4.6: 23.1 (#168)
- QwQ-32B: 22.9 (#171)

| Benchmark | GLM-4.6 | QwQ-32B |
|---|---|---|
| LMArena Hard Prompts | 1440 | 1326 |
| Kagi LLM Benchmark | 47.4% | — |
| CritPt | 1.1% | — |
| Chess Puzzles | — | 5% |
| LiveBench Reasoning | — | 83.5% |
| LiveBench Data Analysis | — | 65% |
| Epoch Capabilities Index | — | 137.6 |
| ForecastBench | — | 58.3 |
| LiveBench | — | 72% |

## Math

- GLM-4.6: 36.4 (#115)
- QwQ-32B: 36.2 (#123)

| Benchmark | GLM-4.6 | QwQ-32B |
|---|---|---|
| LMArena Math | 1431 | 1359 |
| OTIS Mock AIME 2024-2025 | — | 59.2% |
| LiveBench Math | — | 77.8% |
| FrontierMath (Feb 2025 set) | 3.8% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |

## Knowledge

- GLM-4.6: 39.6 (#118)
- QwQ-32B: 37.2 (#150)

| Benchmark | GLM-4.6 | QwQ-32B |
|---|---|---|
| LMArena Expert | 1432 | 1323 |
| GPQA Diamond | — | 65.3% |
| Confabulations | — | 15.6% |
| Vectara Hallucination Rate | 9.5% | — |

## Multilingual

- GLM-4.6: 53.6 (#62)
- QwQ-32B: 45.0 (#171)

| Benchmark | GLM-4.6 | QwQ-32B |
|---|---|---|
| LMArena Non-English | 1426 | 1306 |
| LMArena Chinese | 1500 | 1378 |
| LMArena French | 1459 | 1337 |
| LMArena German | 1447 | 1313 |
| LMArena Japanese | 1394 | 1263 |
| LMArena Korean | 1401 | 1278 |
| LMArena Russian | 1418 | 1297 |
| LMArena Spanish | 1436 | 1353 |

## Instruction Following

- GLM-4.6: 74.5 (#93)
- QwQ-32B: 72.8 (#131)

| Benchmark | GLM-4.6 | QwQ-32B |
|---|---|---|
| LMArena Instruction Following | 1410 | 1297 |
| LiveBench Instruction Following | — | 81.8% |

## Long Context

- GLM-4.6: 43.4 (#91)
- QwQ-32B: 49.0 (#11)

| Benchmark | GLM-4.6 | QwQ-32B |
|---|---|---|
| LMArena Longer Query | 1423 | 1308 |
| Fiction.LiveBench | — | 83.3% |

## Writing & Preference

- GLM-4.6: 61.3 (#86)
- QwQ-32B: 50.9 (#175)

| Benchmark | GLM-4.6 | QwQ-32B |
|---|---|---|
| LMArena Text | 1440 | 1329 |
| LMArena Creative Writing | 1411 | 1288 |
| EQ-Bench Creative Writing | 1411 | 1257 |
| LMArena Multi-Turn | 1427 | 1314 |
| Short-Story Creative Writing | — | 80.2% |
| LiveBench Language | — | 51.4% |

## FAQ

### Is GLM-4.6 better than QwQ-32B?

GLM-4.6 is the stronger model overall, scoring 40.9 to 39.4 on the Noometry Index.

### Is GLM-4.6 or QwQ-32B better for coding?

GLM-4.6 scores higher on coding benchmarks: 39.9 versus 34.9 in the Noometry coding category.

### How many benchmarks do GLM-4.6 and QwQ-32B share?

18 benchmarks have published results for both models. GLM-4.6 has 29 scored results on Noometry and QwQ-32B has 36.