Model comparison
GLM-5.3 vs Qwen2-72B
GLM-5.3 is the stronger model overall, scoring 54.6 to 29.5 on the Noometry Index.
Last verified . 20 shared benchmarks.
Alibaba (Qwen)Summary
- They share 20 benchmarks with published results for both. GLM-5.3 scores higher in 9 categories and Qwen2-72B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GLM-5.3 leads 58.3 to 21.2.
- The biggest single-benchmark swing is WeirdML: 75.4% for GLM-5.3 and 11.3% for Qwen2-72B.
Side by side
| GLM-5.3 | Qwen2-72B | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Alibaba (Qwen) |
| Noometry Index | 54.6 | 29.5 |
| Released | 2026-08-14 | 2024-06-07 |
| Weights | Open | Open |
| Context window | 1M | — |
| Max output | 131K | — |
| Input $ / M tokens | $1.40 | — |
| Output $ / M tokens | $4.40 | — |
| Results tracked | 41 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.3 leads
GLM-5.3: 59.0 (#12), Qwen2-72B: 28.7 (#303)
| Benchmark | GLM-5.3 | Qwen2-72B |
|---|---|---|
| WeirdML | 75.4% | 11.3% |
| LMArena Coding | 1499 | 1196 |
| DeepSWE | 69% | — |
| FrontierCode | 40.1% | — |
| LMArena WebDev | 1623 | — |
| FrontierSWE | 30.2% | — |
| SciCode | 59% | — |
| BigCodeBench Instruct | — | 38.5% |
| BigCodeBench Complete | — | 54% |
| ALE-Bench | 1,317 | — |
Agentic & Tool Use GLM-5.3 leads
GLM-5.3: 36.3 (#34), Qwen2-72B: 16.9 (#138)
| Benchmark | GLM-5.3 | Qwen2-72B |
|---|---|---|
| APEX-Agents | 56.6% | — |
| TheAgentCompany | — | 1.1% |
| METR Time Horizons | — | 29.9% |
| Vending-Bench 2 | 8,164 | — |
Reasoning GLM-5.3 leads
GLM-5.3: 45.2 (#42), Qwen2-72B: 22.2 (#184)
| Benchmark | GLM-5.3 | Qwen2-72B |
|---|---|---|
| LMArena Hard Prompts | 1491 | 1191 |
| Epoch Capabilities Index | 155.56 | 125.27 |
| NYT Connections (extended) | 74.2% | — |
| CritPt | 19.1% | — |
| Chess Puzzles | 21% | — |
| Mystery Game Puzzles | 33% | — |
| DTBench | 87.7% | — |
| LMCA | 55.5% | — |
| Bench to the Future 3 | 0.15 | — |
Math GLM-5.3 leads
GLM-5.3: 62.1 (#27), Qwen2-72B: 27.9 (#232)
| Benchmark | GLM-5.3 | Qwen2-72B |
|---|---|---|
| LMArena Math | 1493 | 1235 |
| FrontierMath (Tiers 1-3) | 68.8% | — |
| FrontierMath Tier 4 | 29.3% | — |
| OTIS Mock AIME 2024-2025 | 91.1% | — |
| ProofBench | 49% | — |
| MATH Level 5 | — | 39.1% |
Knowledge GLM-5.3 leads
GLM-5.3: 58.3 (#34), Qwen2-72B: 21.2 (#270)
| Benchmark | GLM-5.3 | Qwen2-72B |
|---|---|---|
| GPQA Diamond | 90.9% | 40.8% |
| LMArena Expert | 1512 | 1171 |
| SimpleQA Verified | 41% | — |
| MMLU | — | 82.4% |
Multilingual GLM-5.3 leads
GLM-5.3: 55.9 (#27), Qwen2-72B: 36.0 (#239)
| Benchmark | GLM-5.3 | Qwen2-72B |
|---|---|---|
| LMArena Non-English | 1457 | 1176 |
| LMArena Chinese | 1523 | 1240 |
| LMArena French | 1499 | 1169 |
| LMArena German | 1505 | 1151 |
| LMArena Japanese | 1438 | 1111 |
| LMArena Korean | 1478 | 1083 |
| LMArena Russian | 1460 | 1169 |
| LMArena Spanish | 1444 | 1168 |
Instruction Following GLM-5.3 leads
GLM-5.3: 77.8 (#15), Qwen2-72B: 62.0 (#236)
| Benchmark | GLM-5.3 | Qwen2-72B |
|---|---|---|
| LMArena Instruction Following | 1481 | 1181 |
Long Context GLM-5.3 leads
GLM-5.3: 45.5 (#32), Qwen2-72B: 36.1 (#230)
| Benchmark | GLM-5.3 | Qwen2-72B |
|---|---|---|
| LMArena Longer Query | 1485 | 1193 |
Writing & Preference GLM-5.3 leads
GLM-5.3: 75.9 (#6), Qwen2-72B: 41.0 (#236)
| Benchmark | GLM-5.3 | Qwen2-72B |
|---|---|---|
| LMArena Text | 1473 | 1203 |
| LMArena Creative Writing | 1459 | 1181 |
| LMArena Multi-Turn | 1473 | 1195 |
| EQ-Bench Creative Writing | 2075 | — |
Frequently asked questions
Is GLM-5.3 better than Qwen2-72B?
GLM-5.3 is the stronger model overall, scoring 54.6 to 29.5 on the Noometry Index.
Is GLM-5.3 or Qwen2-72B better for coding?
GLM-5.3 scores higher on coding benchmarks: 59.0 versus 28.7 in the Noometry coding category.
How many benchmarks do GLM-5.3 and Qwen2-72B share?
20 benchmarks have published results for both models. GLM-5.3 has 41 scored results on Noometry and Qwen2-72B has 26.