# GLM-5 vs Qwen3 235B-A22B

> GLM-5 is the stronger model overall, scoring 45.9 to 43.2 on the Noometry Index.

- Canonical page: https://noometry.com/compare/glm-5-vs-qwen3-235b-a22b
- Last updated: 2026-10-10
- Shared benchmarks: 32

## Summary

- They share 32 benchmarks with published results for both. GLM-5 scores higher in 6 categories and Qwen3 235B-A22B in 3 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GLM-5 leads 27.6 to 15.8.
- The biggest single-benchmark swing is ARC-AGI-1: 44.7% for GLM-5 and 11% for Qwen3 235B-A22B.
- Qwen3 235B-A22B is cheaper at $0.70 / $2.80 per million input/output tokens, against $1 / $3.20 for GLM-5.
- GLM-5 accepts more context: 205K tokens versus 131K.

## Snapshot

| | GLM-5 | Qwen3 235B-A22B |
|---|---|---|
| Provider | Z.ai (Zhipu) | Alibaba (Qwen) |
| Noometry Index | 45.9 | 43.2 |
| Rank | 59 | 83 |
| Context | 205K | 131K |
| Input $/M | $1 | $0.70 |
| Output $/M | $3.20 | $2.80 |
| Weights | Open | Open |

## Coding

- GLM-5: 49.0 (#47)
- Qwen3 235B-A22B: 43.8 (#72)

| Benchmark | GLM-5 | Qwen3 235B-A22B |
|---|---|---|
| WeirdML | 48.2% | 41% |
| LMArena Coding | 1461 | 1445 |
| SWE-bench Verified | 72.1% | — |
| SWE-bench Verified (bash only) | 72.8% | — |
| Aider Polyglot | — | 59.6% |
| LMArena WebDev | 1434 | — |
| SWE-bench Multilingual | 69.7% | — |
| SciCode | — | 42.4% |
| ALE-Bench | 765.62 | — |

## Agentic & Tool Use

- GLM-5: 30.9 (#65)
- Qwen3 235B-A22B: 33.7 (#48)

| Benchmark | GLM-5 | Qwen3 235B-A22B |
|---|---|---|
| Vending-Bench 2 | 4,432 | -11.34 |
| Terminal-Bench | 52.4% | — |
| Berkeley Function Calling Leaderboard | — | 52.1% |
| τ²-bench Airline | 82.5% | — |
| τ²-bench Banking | 9.8% | — |
| τ²-bench Retail | 73.7% | — |
| τ²-bench Telecom | 86.8% | — |

## Reasoning

- GLM-5: 27.6 (#100)
- Qwen3 235B-A22B: 15.8 (#302)

| Benchmark | GLM-5 | Qwen3 235B-A22B |
|---|---|---|
| ARC-AGI-2 | 4.9% | 1.3% |
| SimpleBench | 53.2% | 31% |
| Kagi LLM Benchmark | 75% | 69.4% |
| ARC-AGI-1 | 44.7% | 11% |
| Chess Puzzles | 10% | 12% |
| LMArena Hard Prompts | 1451 | 1433 |
| Epoch Capabilities Index | 145.84 | 143.87 |
| ForecastBench | 61 | 59.7 |
| NYT Connections (extended) | 74.8% | — |
| CritPt | — | 0% |
| Mystery Game Puzzles | — | 9% |
| DTBench | — | 80.3% |
| LMCA | — | 29.3% |

## Math

- GLM-5: 44.8 (#66)
- Qwen3 235B-A22B: 48.5 (#51)

| Benchmark | GLM-5 | Qwen3 235B-A22B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 80% | 86.7% |
| LMArena Math | 1440 | 1431 |
| FrontierMath (Feb 2025 set) | 16.4% | 8.5% |
| FrontierMath Tier 4 (v1) | 2.1% | 0% |
| MathArena Final-Answer Competitions | 65.7% | — |
| Omni-MATH | — | 71.8% |
| MATH Level 5 | — | 68.9% |

## Knowledge

- GLM-5: 52.2 (#59)
- Qwen3 235B-A22B: 49.5 (#67)

| Benchmark | GLM-5 | Qwen3 235B-A22B |
|---|---|---|
| GPQA Diamond | 87.8% | 80.1% |
| Vectara Hallucination Rate | 10.1% | 9.3% |
| LMArena Expert | 1454 | 1460 |
| SimpleQA Verified | — | 40.4% |
| MMLU-Pro | — | 84.4% |
| Confabulations | — | 15.6% |
| GPQA (HELM) | — | 72.7% |

## Multilingual

- GLM-5: 53.9 (#57)
- Qwen3 235B-A22B: 52.5 (#84)

| Benchmark | GLM-5 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Non-English | 1430 | 1410 |
| LMArena Chinese | 1511 | 1481 |
| LMArena French | 1455 | 1445 |
| LMArena German | 1445 | 1432 |
| LMArena Japanese | 1416 | 1400 |
| LMArena Korean | 1421 | 1391 |
| LMArena Russian | 1436 | 1411 |
| LMArena Spanish | 1452 | 1429 |

## Instruction Following

- GLM-5: 75.4 (#64)
- Qwen3 235B-A22B: 72.8 (#132)

| Benchmark | GLM-5 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Instruction Following | 1427 | 1408 |
| IFEval | — | 83.5% |

## Long Context

- GLM-5: 44.6 (#58)
- Qwen3 235B-A22B: 46.1 (#25)

| Benchmark | GLM-5 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Longer Query | 1446 | 1426 |
| Fiction.LiveBench | — | 75% |
| CL-bench | 18.7% | — |

## Writing & Preference

- GLM-5: 66.2 (#38)
- Qwen3 235B-A22B: 59.8 (#103)

| Benchmark | GLM-5 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Text | 1446 | 1419 |
| LMArena Creative Writing | 1438 | 1384 |
| EQ-Bench Creative Writing | 1601 | 1366 |
| LMArena Multi-Turn | 1456 | 1432 |
| Short-Story Creative Writing | — | 83% |
| WildBench | — | 86.6% |

## FAQ

### Is GLM-5 better than Qwen3 235B-A22B?

GLM-5 is the stronger model overall, scoring 45.9 to 43.2 on the Noometry Index.

### Which is cheaper, GLM-5 or Qwen3 235B-A22B?

Qwen3 235B-A22B is cheaper. It lists at $0.70 per million input tokens and $2.80 per million output tokens; GLM-5 lists at $1 and $3.20.

### Is GLM-5 or Qwen3 235B-A22B better for coding?

GLM-5 scores higher on coding benchmarks: 49.0 versus 43.8 in the Noometry coding category.

### Which has the bigger context window?

GLM-5 does, with 205K tokens against 131K.

### How many benchmarks do GLM-5 and Qwen3 235B-A22B share?

32 benchmarks have published results for both models. GLM-5 has 45 scored results on Noometry and Qwen3 235B-A22B has 49.