Model comparison
GLM-5.3 vs Qwen3.6 Max Preview
GLM-5.3 is the stronger model overall, scoring 54.8 to 51.5 on the Noometry Index.
Last verified . 25 shared benchmarks.
Alibaba (Qwen)Summary
- They share 25 benchmarks with published results for both. GLM-5.3 scores higher in 8 categories and Qwen3.6 Max Preview in 0 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GLM-5.3 leads 75.7 to 63.8.
- The biggest single-benchmark swing is Mystery Game Puzzles: 33% for GLM-5.3 and 19% for Qwen3.6 Max Preview.
- GLM-5.3 is cheaper at $1.40 / $4.40 per million input/output tokens, against $1.30 / $7.80 for Qwen3.6 Max Preview.
- GLM-5.3 accepts more context: 1M tokens versus 262K.
- GLM-5.3 has downloadable open weights; the other is API-only.
Side by side
| GLM-5.3 | Qwen3.6 Max Preview | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Alibaba (Qwen) |
| Noometry Index | 54.8 | 51.5 |
| Released | 2026-08-14 | 2026-04-20 |
| Weights | Open | Proprietary |
| Context window | 1M | 262K |
| Max output | 131K | 66K |
| Input $ / M tokens | $1.40 | $1.30 |
| Output $ / M tokens | $4.40 | $7.80 |
| Results tracked | 42 | 29 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.3 leads
GLM-5.3: 59.5 (#14), Qwen3.6 Max Preview: 48.7 (#54)
| Benchmark | GLM-5.3 | Qwen3.6 Max Preview |
|---|---|---|
| LMArena WebDev | 1622 | 1482 |
| LMArena Coding | 1496 | 1471 |
| SWE-bench Verified | — | 76.7% |
| DeepSWE | 69% | — |
| FrontierCode | 40.1% | — |
| CursorBench | 42.6% | — |
| FrontierSWE | 30.2% | — |
| SciCode | 59% | — |
| WeirdML | 75.4% | — |
| ALE-Bench | 1,317 | — |
Agentic & Tool Use Not comparable
GLM-5.3: 36.4 (#38), Qwen3.6 Max Preview: —
| Benchmark | GLM-5.3 | Qwen3.6 Max Preview |
|---|---|---|
| Vending-Bench 2 | 8,164 | 4,254 |
| APEX-Agents | 56.6% | — |
Reasoning GLM-5.3 leads
GLM-5.3: 46.1 (#46), Qwen3.6 Max Preview: 41.7 (#53)
| Benchmark | GLM-5.3 | Qwen3.6 Max Preview |
|---|---|---|
| NYT Connections (extended) | 74.2% | 74.1% |
| Chess Puzzles | 21% | 20% |
| LMArena Hard Prompts | 1489 | 1457 |
| Mystery Game Puzzles | 33% | 19% |
| DTBench | 87.7% | 87.2% |
| LMCA | 55.5% | 42.5% |
| Epoch Capabilities Index | 155.61 | 149.24 |
| SimpleBench | — | 63% |
| CritPt | 19.1% | — |
| Bench to the Future 3 | 0.15 | — |
Math GLM-5.3 leads
GLM-5.3: 62.3 (#33), Qwen3.6 Max Preview: 54.1 (#46)
| Benchmark | GLM-5.3 | Qwen3.6 Max Preview |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 91.1% | 91.1% |
| LMArena Math | 1489 | 1465 |
| FrontierMath (Tiers 1-3) | 68.8% | — |
| FrontierMath Tier 4 | 29.3% | — |
| ProofBench | 49% | — |
| FrontierMath (Feb 2025 set) | — | 23.1% |
| FrontierMath Tier 4 (v1) | — | 4.2% |
Knowledge Too close to call
GLM-5.3: 58.3 (#37), Qwen3.6 Max Preview: 57.6 (#39)
| Benchmark | GLM-5.3 | Qwen3.6 Max Preview |
|---|---|---|
| GPQA Diamond | 90.9% | 87.4% |
| SimpleQA Verified | 41% | 52% |
| LMArena Expert | 1516 | 1478 |
Multilingual GLM-5.3 leads
GLM-5.3: 55.7 (#28), Qwen3.6 Max Preview: 54.2 (#48)
| Benchmark | GLM-5.3 | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Non-English | 1457 | 1437 |
| LMArena Chinese | 1528 | 1487 |
| LMArena French | 1499 | 1449 |
| LMArena Russian | 1463 | 1445 |
| LMArena Spanish | 1460 | 1454 |
| LMArena German | 1499 | — |
| LMArena Japanese | 1453 | — |
| LMArena Korean | 1472 | — |
Instruction Following GLM-5.3 leads
GLM-5.3: 77.5 (#23), Qwen3.6 Max Preview: 75.7 (#55)
| Benchmark | GLM-5.3 | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Instruction Following | 1477 | 1438 |
Long Context Too close to call
GLM-5.3: 45.4 (#41), Qwen3.6 Max Preview: 44.6 (#61)
| Benchmark | GLM-5.3 | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Longer Query | 1482 | 1457 |
Writing & Preference GLM-5.3 leads
GLM-5.3: 75.7 (#6), Qwen3.6 Max Preview: 63.8 (#60)
| Benchmark | GLM-5.3 | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Text | 1471 | 1447 |
| LMArena Creative Writing | 1457 | 1435 |
| LMArena Multi-Turn | 1472 | 1456 |
| EQ-Bench Creative Writing | 2075 | — |
Frequently asked questions
Is GLM-5.3 better than Qwen3.6 Max Preview?
GLM-5.3 is the stronger model overall, scoring 54.8 to 51.5 on the Noometry Index.
Which is cheaper, GLM-5.3 or Qwen3.6 Max Preview?
GLM-5.3 is cheaper. It lists at $1.40 per million input tokens and $4.40 per million output tokens; Qwen3.6 Max Preview lists at $1.30 and $7.80.
Is GLM-5.3 or Qwen3.6 Max Preview better for coding?
GLM-5.3 scores higher on coding benchmarks: 59.5 versus 48.7 in the Noometry coding category.
Which has the bigger context window?
GLM-5.3 does, with 1M tokens against 262K.
How many benchmarks do GLM-5.3 and Qwen3.6 Max Preview share?
25 benchmarks have published results for both models. GLM-5.3 has 42 scored results on Noometry and Qwen3.6 Max Preview has 29.