Model comparison
GLM-4.6 vs Qwen2-72B
GLM-4.6 is the stronger model overall, scoring 41.4 to 30.0 on the Noometry Index.
Last verified . 17 shared benchmarks.
Alibaba (Qwen)Summary
- They share 17 benchmarks with published results for both. GLM-4.6 scores higher in 9 categories and Qwen2-72B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GLM-4.6 leads 61.1 to 40.8.
Side by side
| GLM-4.6 | Qwen2-72B | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Alibaba (Qwen) |
| Noometry Index | 41.4 | 30.0 |
| Released | 2025-09-30 | 2024-06-07 |
| Weights | Open | Open |
| Context window | 205K | — |
| Max output | 131K | — |
| Input $ / M tokens | $0.60 | — |
| Output $ / M tokens | $2.20 | — |
| Results tracked | 29 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-4.6 leads
GLM-4.6: 40.1 (#148), Qwen2-72B: 29.1 (#310)
| Benchmark | GLM-4.6 | Qwen2-72B |
|---|---|---|
| LMArena Coding | 1449 | 1196 |
| SWE-bench Verified (bash only) | 55.4% | — |
| LMArena WebDev | 1340 | — |
| SciCode | 38.4% | — |
| WeirdML | — | 11.3% |
| BigCodeBench Instruct | — | 38.5% |
| BigCodeBench Complete | — | 54% |
| ALE-Bench | 340.82 | — |
Agentic & Tool Use GLM-4.6 leads
GLM-4.6: 32.3 (#66), Qwen2-72B: 17.0 (#146)
| Benchmark | GLM-4.6 | Qwen2-72B |
|---|---|---|
| Terminal-Bench | 24.5% | — |
| Berkeley Function Calling Leaderboard | 72.4% | — |
| TheAgentCompany | — | 1.1% |
| METR Time Horizons | — | 29.9% |
Reasoning Too close to call
GLM-4.6: 23.7 (#172), Qwen2-72B: 23.2 (#181)
| Benchmark | GLM-4.6 | Qwen2-72B |
|---|---|---|
| LMArena Hard Prompts | 1440 | 1191 |
| Kagi LLM Benchmark | 47.4% | — |
| CritPt | 1.1% | — |
| Epoch Capabilities Index | — | 125.28 |
Math GLM-4.6 leads
GLM-4.6: 39.1 (#111), Qwen2-72B: 30.2 (#236)
| Benchmark | GLM-4.6 | Qwen2-72B |
|---|---|---|
| LMArena Math | 1432 | 1235 |
| MATH Level 5 | — | 39.1% |
| FrontierMath (Feb 2025 set) | 3.8% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge GLM-4.6 leads
GLM-4.6: 40.2 (#124), Qwen2-72B: 21.2 (#275)
| Benchmark | GLM-4.6 | Qwen2-72B |
|---|---|---|
| LMArena Expert | 1431 | 1171 |
| GPQA Diamond | — | 40.8% |
| Vectara Hallucination Rate | 9.5% | — |
| MMLU | — | 82.4% |
Multilingual GLM-4.6 leads
GLM-4.6: 53.5 (#66), Qwen2-72B: 35.9 (#244)
| Benchmark | GLM-4.6 | Qwen2-72B |
|---|---|---|
| LMArena Non-English | 1426 | 1176 |
| LMArena Chinese | 1499 | 1240 |
| LMArena French | 1459 | 1170 |
| LMArena German | 1447 | 1151 |
| LMArena Japanese | 1393 | 1111 |
| LMArena Korean | 1400 | 1083 |
| LMArena Russian | 1419 | 1169 |
| LMArena Spanish | 1436 | 1169 |
Instruction Following GLM-4.6 leads
GLM-4.6: 74.3 (#98), Qwen2-72B: 61.7 (#241)
| Benchmark | GLM-4.6 | Qwen2-72B |
|---|---|---|
| LMArena Instruction Following | 1410 | 1181 |
Long Context GLM-4.6 leads
GLM-4.6: 43.4 (#94), Qwen2-72B: 36.1 (#235)
| Benchmark | GLM-4.6 | Qwen2-72B |
|---|---|---|
| LMArena Longer Query | 1422 | 1192 |
Writing & Preference GLM-4.6 leads
GLM-4.6: 61.1 (#90), Qwen2-72B: 40.8 (#241)
| Benchmark | GLM-4.6 | Qwen2-72B |
|---|---|---|
| LMArena Text | 1440 | 1203 |
| LMArena Creative Writing | 1411 | 1181 |
| LMArena Multi-Turn | 1427 | 1196 |
| EQ-Bench Creative Writing | 1411 | — |
Frequently asked questions
Is GLM-4.6 better than Qwen2-72B?
GLM-4.6 is the stronger model overall, scoring 41.4 to 30.0 on the Noometry Index.
Is GLM-4.6 or Qwen2-72B better for coding?
GLM-4.6 scores higher on coding benchmarks: 40.1 versus 29.1 in the Noometry coding category.
How many benchmarks do GLM-4.6 and Qwen2-72B share?
17 benchmarks have published results for both models. GLM-4.6 has 29 scored results on Noometry and Qwen2-72B has 26.