Model comparison
GLM-5.3 vs Qwen3.5-4B
GLM-5.3 has enough public results to be ranked (#25); Qwen3.5-4B does not yet, so treat this comparison as directional.
Last verified . 1 shared benchmarks.
Alibaba (Qwen)Summary
- They share 1 benchmark with published results for both. GLM-5.3 scores higher in 1 category and Qwen3.5-4B in 0 categories; one gap is clear of the uncertainty.
- The widest gap is in math, where GLM-5.3 leads 62.1 to 33.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 91.1% for GLM-5.3 and 55.8% for Qwen3.5-4B.
Side by side
| GLM-5.3 | Qwen3.5-4B | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Alibaba (Qwen) |
| Noometry Index | 54.6 | 36.2 |
| Released | 2026-08-14 | 2026-02-24 |
| Weights | Open | Open |
| Context window | 1M | — |
| Max output | 131K | — |
| Input $ / M tokens | $1.40 | — |
| Output $ / M tokens | $4.40 | — |
| Results tracked | 41 | 1 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
GLM-5.3: 59.0 (#12), Qwen3.5-4B: —
| Benchmark | GLM-5.3 | Qwen3.5-4B |
|---|---|---|
| DeepSWE | 69% | — |
| FrontierCode | 40.1% | — |
| LMArena WebDev | 1623 | — |
| FrontierSWE | 30.2% | — |
| SciCode | 59% | — |
| WeirdML | 75.4% | — |
| LMArena Coding | 1499 | — |
| ALE-Bench | 1,317 | — |
Agentic & Tool Use Not comparable
GLM-5.3: 36.3 (#34), Qwen3.5-4B: —
| Benchmark | GLM-5.3 | Qwen3.5-4B |
|---|---|---|
| APEX-Agents | 56.6% | — |
| Vending-Bench 2 | 8,164 | — |
Reasoning Not comparable
GLM-5.3: 45.2 (#42), Qwen3.5-4B: —
| Benchmark | GLM-5.3 | Qwen3.5-4B |
|---|---|---|
| NYT Connections (extended) | 74.2% | — |
| CritPt | 19.1% | — |
| Chess Puzzles | 21% | — |
| LMArena Hard Prompts | 1491 | — |
| Mystery Game Puzzles | 33% | — |
| DTBench | 87.7% | — |
| LMCA | 55.5% | — |
| Bench to the Future 3 | 0.15 | — |
| Epoch Capabilities Index | 155.56 | — |
Math GLM-5.3 leads
GLM-5.3: 62.1 (#27), Qwen3.5-4B: 33.2
| Benchmark | GLM-5.3 | Qwen3.5-4B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 91.1% | 55.8% |
| FrontierMath (Tiers 1-3) | 68.8% | — |
| FrontierMath Tier 4 | 29.3% | — |
| ProofBench | 49% | — |
| LMArena Math | 1493 | — |
Knowledge Not comparable
GLM-5.3: 58.3 (#34), Qwen3.5-4B: —
| Benchmark | GLM-5.3 | Qwen3.5-4B |
|---|---|---|
| GPQA Diamond | 90.9% | — |
| SimpleQA Verified | 41% | — |
| LMArena Expert | 1512 | — |
Multilingual Not comparable
GLM-5.3: 55.9 (#27), Qwen3.5-4B: —
| Benchmark | GLM-5.3 | Qwen3.5-4B |
|---|---|---|
| LMArena Non-English | 1457 | — |
| LMArena Chinese | 1523 | — |
| LMArena French | 1499 | — |
| LMArena German | 1505 | — |
| LMArena Japanese | 1438 | — |
| LMArena Korean | 1478 | — |
| LMArena Russian | 1460 | — |
| LMArena Spanish | 1444 | — |
Instruction Following Not comparable
GLM-5.3: 77.8 (#15), Qwen3.5-4B: —
| Benchmark | GLM-5.3 | Qwen3.5-4B |
|---|---|---|
| LMArena Instruction Following | 1481 | — |
Long Context Not comparable
GLM-5.3: 45.5 (#32), Qwen3.5-4B: —
| Benchmark | GLM-5.3 | Qwen3.5-4B |
|---|---|---|
| LMArena Longer Query | 1485 | — |
Writing & Preference Not comparable
GLM-5.3: 75.9 (#6), Qwen3.5-4B: —
| Benchmark | GLM-5.3 | Qwen3.5-4B |
|---|---|---|
| LMArena Text | 1473 | — |
| LMArena Creative Writing | 1459 | — |
| EQ-Bench Creative Writing | 2075 | — |
| LMArena Multi-Turn | 1473 | — |
Frequently asked questions
Is GLM-5.3 better than Qwen3.5-4B?
GLM-5.3 has enough public results to be ranked (#25); Qwen3.5-4B does not yet, so treat this comparison as directional.
How many benchmarks do GLM-5.3 and Qwen3.5-4B share?
1 benchmark has published results for both models. GLM-5.3 has 41 scored results on Noometry and Qwen3.5-4B has 1.