Model comparison
GLM-5.3-Flash vs Qwen2.5-Coder-32B
GLM-5.3-Flash is the stronger model overall, scoring 51.8 to 33.4 on the Noometry Index.
Last verified . 13 shared benchmarks.
Alibaba (Qwen)Summary
- They share 13 benchmarks with published results for both. GLM-5.3-Flash scores higher in 8 categories and Qwen2.5-Coder-32B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in coding, where GLM-5.3-Flash leads 53.1 to 22.6.
- GLM-5.3-Flash is cheaper at $0.15 / $0.50 per million input/output tokens, against $0.66 / $1 for Qwen2.5-Coder-32B.
- GLM-5.3-Flash accepts more context: 1M tokens versus 33K.
Side by side
| GLM-5.3-Flash | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Alibaba (Qwen) |
| Noometry Index | 51.8 | 33.4 |
| Released | 2026-08-20 | 2024-09-18 |
| Weights | Open | Open |
| Context window | 1M | 33K |
| Max output | 131K | 29K |
| Input $ / M tokens | $0.15 | $0.66 |
| Output $ / M tokens | $0.50 | $1 |
| Results tracked | 40 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.3-Flash leads
GLM-5.3-Flash: 53.1 (#31), Qwen2.5-Coder-32B: 22.6 (#333)
| Benchmark | GLM-5.3-Flash | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Coding | 1508 | 1276 |
| DeepSWE | 63.4% | — |
| FrontierCode | 31.8% | — |
| SWE-bench Verified (bash only) | — | 9% |
| Aider Polyglot | — | 16.4% |
| CursorBench | 36.8% | — |
| LMArena WebDev | 1609 | — |
| FrontierSWE | 18.1% | — |
| SciCode | 51.6% | — |
| BigCodeBench Instruct | — | 49% |
| LiveBench Coding | — | 56.9% |
| BigCodeBench Complete | — | 58% |
| ALE-Bench | 303.55 | — |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 77% |
Agentic & Tool Use Not comparable
GLM-5.3-Flash: 34.2 (#47), Qwen2.5-Coder-32B: —
| Benchmark | GLM-5.3-Flash | Qwen2.5-Coder-32B |
|---|---|---|
| APEX-Agents | 52.8% | — |
| GDP.pdf | 14% | — |
Reasoning GLM-5.3-Flash leads
GLM-5.3-Flash: 48.0 (#42), Qwen2.5-Coder-32B: 21.2 (#225)
| Benchmark | GLM-5.3-Flash | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Hard Prompts | 1491 | 1251 |
| Epoch Capabilities Index | 151.88 | 119.49 |
| ARC-AGI-2 | 65.8% | — |
| ARC-AGI-1 | 91% | — |
| CritPt | 15.4% | — |
| Chess Puzzles | 14% | — |
| LiveBench Reasoning | — | 42.1% |
| Mystery Game Puzzles | 8% | — |
| LiveBench Data Analysis | — | 49.9% |
| Surface Evolver Bench | 52.5% | — |
| Bench to the Future 3 | 0.15 | — |
| HellaSwag | — | 83% |
| LiveBench | — | 46.2% |
| WinoGrande | — | 80.8% |
Math GLM-5.3-Flash leads
GLM-5.3-Flash: 53.3 (#47), Qwen2.5-Coder-32B: 33.3 (#204)
| Benchmark | GLM-5.3-Flash | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Math | 1500 | 1251 |
| FrontierMath (Tiers 1-3) | 55.8% | — |
| FrontierMath Tier 4 | 17.1% | — |
| OTIS Mock AIME 2024-2025 | 93.9% | — |
| ProofBench | 21% | — |
| LiveBench Math | — | 46.6% |
| GSM8K | — | 93% |
Knowledge GLM-5.3-Flash leads
GLM-5.3-Flash: 58.4 (#36), Qwen2.5-Coder-32B: 33.4 (#203)
| Benchmark | GLM-5.3-Flash | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1513 | 1221 |
| GPQA Diamond | 90.2% | — |
| ARC (AI2) Challenge | — | 70.5% |
| MMLU | — | 79.1% |
Multimodal Not comparable
GLM-5.3-Flash: 42.8 (#27), Qwen2.5-Coder-32B: —
| Benchmark | GLM-5.3-Flash | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Vision | 1296 | — |
Multilingual GLM-5.3-Flash leads
GLM-5.3-Flash: 56.0 (#25), Qwen2.5-Coder-32B: 37.8 (#235)
| Benchmark | GLM-5.3-Flash | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1462 | 1205 |
| LMArena Chinese | 1527 | 1222 |
| LMArena Russian | 1469 | 1228 |
| LMArena French | 1496 | — |
| LMArena German | 1470 | — |
| LMArena Japanese | 1429 | — |
| LMArena Korean | 1446 | — |
| LMArena Spanish | 1471 | — |
Instruction Following GLM-5.3-Flash leads
GLM-5.3-Flash: 77.5 (#20), Qwen2.5-Coder-32B: 61.4 (#245)
| Benchmark | GLM-5.3-Flash | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Instruction Following | 1478 | 1223 |
| LiveBench Instruction Following | — | 58.7% |
Long Context GLM-5.3-Flash leads
GLM-5.3-Flash: 45.4 (#39), Qwen2.5-Coder-32B: 38.0 (#208)
| Benchmark | GLM-5.3-Flash | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1482 | 1251 |
Writing & Preference GLM-5.3-Flash leads
GLM-5.3-Flash: 65.3 (#50), Qwen2.5-Coder-32B: 41.6 (#240)
| Benchmark | GLM-5.3-Flash | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1471 | 1230 |
| LMArena Creative Writing | 1442 | 1174 |
| LMArena Multi-Turn | 1467 | 1222 |
| LiveBench Language | — | 23.3% |
Frequently asked questions
Is GLM-5.3-Flash better than Qwen2.5-Coder-32B?
GLM-5.3-Flash is the stronger model overall, scoring 51.8 to 33.4 on the Noometry Index.
Which is cheaper, GLM-5.3-Flash or Qwen2.5-Coder-32B?
GLM-5.3-Flash is cheaper. It lists at $0.15 per million input tokens and $0.50 per million output tokens; Qwen2.5-Coder-32B lists at $0.66 and $1.
Is GLM-5.3-Flash or Qwen2.5-Coder-32B better for coding?
GLM-5.3-Flash scores higher on coding benchmarks: 53.1 versus 22.6 in the Noometry coding category.
Which has the bigger context window?
GLM-5.3-Flash does, with 1M tokens against 33K.
How many benchmarks do GLM-5.3-Flash and Qwen2.5-Coder-32B share?
13 benchmarks have published results for both models. GLM-5.3-Flash has 40 scored results on Noometry and Qwen2.5-Coder-32B has 31.