Model comparison
GLM-5.2 vs Qwen3 Max
GLM-5.2 is the stronger model overall, scoring 51.1 to 43.7 on the Noometry Index.
Last verified . 30 shared benchmarks.
Alibaba (Qwen)Summary
- They share 30 benchmarks with published results for both. GLM-5.2 scores higher in 8 categories and Qwen3 Max in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GLM-5.2 leads 42.3 to 22.6.
- The biggest single-benchmark swing is NYT Connections (extended): 74.3% for GLM-5.2 and 30.1% for Qwen3 Max.
- GLM-5.2 is cheaper at $1.40 / $4.40 per million input/output tokens, against $1.20 / $6 for Qwen3 Max.
- GLM-5.2 accepts more context: 1M tokens versus 262K.
- GLM-5.2 has downloadable open weights; the other is API-only.
Side by side
| GLM-5.2 | Qwen3 Max | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Alibaba (Qwen) |
| Noometry Index | 51.1 | 43.7 |
| Released | 2026-06-13 | 2025-09-23 |
| Weights | Open | Proprietary |
| Context window | 1M | 262K |
| Max output | 131K | 66K |
| Input $ / M tokens | $1.40 | $1.20 |
| Output $ / M tokens | $4.40 | $6 |
| Results tracked | 51 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.2 leads
GLM-5.2: 51.3 (#41), Qwen3 Max: 43.0 (#93)
| Benchmark | GLM-5.2 | Qwen3 Max |
|---|---|---|
| LMArena Coding | 1485 | 1456 |
| ALE-Bench | 1,047 | 370.45 |
| SWE-bench Verified | 78.7% | — |
| DeepSWE | 43.8% | — |
| FrontierCode | 24.5% | — |
| LMArena WebDev | 1603 | — |
| SciCode | 50.5% | — |
| WeirdML | 70.1% | — |
Agentic & Tool Use Not comparable
GLM-5.2: 32.4 (#63), Qwen3 Max: —
| Benchmark | GLM-5.2 | Qwen3 Max |
|---|---|---|
| Vending-Bench 2 | 8,314 | 71.56 |
| APEX-Agents | 45.2% | — |
| τ²-bench Banking | 37.1% | — |
| PostTrainBench | 31.7% | — |
| GBAEval | 0% | — |
Reasoning GLM-5.2 leads
GLM-5.2: 42.3 (#52), Qwen3 Max: 22.6 (#190)
| Benchmark | GLM-5.2 | Qwen3 Max |
|---|---|---|
| Kagi LLM Benchmark | 62.6% | 72.5% |
| NYT Connections (extended) | 74.3% | 30.1% |
| Chess Puzzles | 21% | 4% |
| LMArena Hard Prompts | 1480 | 1448 |
| Mystery Game Puzzles | 19% | 5% |
| DTBench | 93.6% | 82.1% |
| LMCA | 45.8% | 28.3% |
| Epoch Capabilities Index | 151.78 | 142.38 |
| ARC-AGI-2 | 22.8% | — |
| SimpleBench | 58.8% | — |
| ARC-AGI-1 | 77% | — |
| CritPt | 20.9% | — |
| EBR-Bench | 9.5% | — |
| Surface Evolver Bench | 55.6% | — |
Math GLM-5.2 leads
GLM-5.2: 55.7 (#43), Qwen3 Max: 38.7 (#131)
| Benchmark | GLM-5.2 | Qwen3 Max |
|---|---|---|
| FrontierMath (Tiers 1-3) | 59.2% | 18.9% |
| OTIS Mock AIME 2024-2025 | 86.4% | 73.3% |
| LMArena Math | 1482 | 1446 |
| FrontierMath Tier 4 | 29.3% | — |
| MathArena Final-Answer Competitions | 67.6% | — |
| ProofBench | 35% | — |
| MATH Level 5 | — | 97.1% |
Knowledge GLM-5.2 leads
GLM-5.2: 57.1 (#40), Qwen3 Max: 48.1 (#78)
| Benchmark | GLM-5.2 | Qwen3 Max |
|---|---|---|
| GPQA Diamond | 91.9% | 72.6% |
| SimpleQA Verified | 34.2% | 48.7% |
| LMArena Expert | 1486 | 1455 |
Multilingual GLM-5.2 leads
GLM-5.2: 55.8 (#26), Qwen3 Max: 53.7 (#62)
| Benchmark | GLM-5.2 | Qwen3 Max |
|---|---|---|
| LMArena Non-English | 1459 | 1429 |
| LMArena Chinese | 1519 | 1478 |
| LMArena French | 1479 | 1449 |
| LMArena German | 1468 | 1463 |
| LMArena Japanese | 1451 | 1397 |
| LMArena Korean | 1445 | 1399 |
| LMArena Russian | 1466 | 1428 |
| LMArena Spanish | 1477 | 1462 |
Instruction Following GLM-5.2 leads
GLM-5.2: 76.9 (#34), Qwen3 Max: 74.8 (#87)
| Benchmark | GLM-5.2 | Qwen3 Max |
|---|---|---|
| LMArena Instruction Following | 1465 | 1419 |
Long Context GLM-5.2 leads
GLM-5.2: 45.3 (#43), Qwen3 Max: 41.6 (#134)
| Benchmark | GLM-5.2 | Qwen3 Max |
|---|---|---|
| LMArena Longer Query | 1479 | 1438 |
| Fiction.LiveBench | — | 66.7% |
| CL-bench | — | 14.5% |
Writing & Preference GLM-5.2 leads
GLM-5.2: 70.4 (#21), Qwen3 Max: 62.4 (#76)
| Benchmark | GLM-5.2 | Qwen3 Max |
|---|---|---|
| LMArena Text | 1470 | 1439 |
| LMArena Creative Writing | 1462 | 1402 |
| LMArena Multi-Turn | 1469 | 1446 |
| EQ-Bench Creative Writing | 1757 | — |
| EQ-Bench 4 | 1222 | — |
Frequently asked questions
Is GLM-5.2 better than Qwen3 Max?
GLM-5.2 is the stronger model overall, scoring 51.1 to 43.7 on the Noometry Index.
Which is cheaper, GLM-5.2 or Qwen3 Max?
GLM-5.2 is cheaper. It lists at $1.40 per million input tokens and $4.40 per million output tokens; Qwen3 Max lists at $1.20 and $6.
Is GLM-5.2 or Qwen3 Max better for coding?
GLM-5.2 scores higher on coding benchmarks: 51.3 versus 43.0 in the Noometry coding category.
Which has the bigger context window?
GLM-5.2 does, with 1M tokens against 262K.
How many benchmarks do GLM-5.2 and Qwen3 Max share?
30 benchmarks have published results for both models. GLM-5.2 has 51 scored results on Noometry and Qwen3 Max has 33.