Model comparison
GLM-5.3-Flash vs Qwen3.5 27B
GLM-5.3-Flash is the stronger model overall, scoring 51.8 to 41.9 on the Noometry Index.
Last verified . 20 shared benchmarks.
Summary
- They share 20 benchmarks with published results for both. GLM-5.3-Flash scores higher in 9 categories and Qwen3.5 27B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GLM-5.3-Flash leads 48.0 to 27.5.
- GLM-5.3-Flash is cheaper at $0.15 / $0.50 per million input/output tokens, against $0.30 / $2.40 for Qwen3.5 27B.
- GLM-5.3-Flash accepts more context: 1M tokens versus 262K.
Side by side
| GLM-5.3-Flash | Qwen3.5 27B | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Alibaba (Qwen) |
| Noometry Index | 51.8 | 41.9 |
| Released | 2026-08-20 | 2026-02-23 |
| Weights | Open | Open |
| Context window | 1M | 262K |
| Max output | 131K | 66K |
| Input $ / M tokens | $0.15 | $0.30 |
| Output $ / M tokens | $0.50 | $2.40 |
| Results tracked | 40 | 28 |
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), Qwen3.5 27B: 38.9 (#168)
| Benchmark | GLM-5.3-Flash | Qwen3.5 27B |
|---|---|---|
| LMArena WebDev | 1609 | 1358 |
| LMArena Coding | 1508 | 1427 |
| ALE-Bench | 303.55 | 349.45 |
| DeepSWE | 63.4% | — |
| FrontierCode | 31.8% | — |
| CursorBench | 36.8% | — |
| FrontierSWE | 18.1% | — |
| SciCode | 51.6% | — |
| WeirdML | — | 39.5% |
Agentic & Tool Use Not comparable
GLM-5.3-Flash: 34.2 (#47), Qwen3.5 27B: —
| Benchmark | GLM-5.3-Flash | Qwen3.5 27B |
|---|---|---|
| APEX-Agents | 52.8% | — |
| GDP.pdf | 14% | — |
| Vending-Bench 2 | — | 201.98 |
Reasoning GLM-5.3-Flash leads
GLM-5.3-Flash: 48.0 (#42), Qwen3.5 27B: 27.5 (#117)
| Benchmark | GLM-5.3-Flash | Qwen3.5 27B |
|---|---|---|
| LMArena Hard Prompts | 1491 | 1414 |
| ARC-AGI-2 | 65.8% | — |
| NYT Connections (extended) | — | 47.9% |
| ARC-AGI-1 | 91% | — |
| CritPt | 15.4% | — |
| Chess Puzzles | 14% | — |
| Thematic Generalization | — | 45.5% |
| Mystery Game Puzzles | 8% | — |
| DTBench | — | 82.4% |
| LMCA | — | 34% |
| Surface Evolver Bench | 52.5% | — |
| Bench to the Future 3 | 0.15 | — |
| Epoch Capabilities Index | 151.88 | — |
Math GLM-5.3-Flash leads
GLM-5.3-Flash: 53.3 (#47), Qwen3.5 27B: 38.8 (#127)
| Benchmark | GLM-5.3-Flash | Qwen3.5 27B |
|---|---|---|
| LMArena Math | 1500 | 1429 |
| FrontierMath (Tiers 1-3) | 55.8% | — |
| FrontierMath Tier 4 | 17.1% | — |
| MathArena Final-Answer Competitions | — | 56.7% |
| OTIS Mock AIME 2024-2025 | 93.9% | — |
| ProofBench | 21% | — |
Knowledge GLM-5.3-Flash leads
GLM-5.3-Flash: 58.4 (#36), Qwen3.5 27B: 38.0 (#150)
| Benchmark | GLM-5.3-Flash | Qwen3.5 27B |
|---|---|---|
| LMArena Expert | 1513 | 1428 |
| GPQA Diamond | 90.2% | — |
| Vectara Hallucination Rate | — | 12.1% |
Multimodal GLM-5.3-Flash leads
GLM-5.3-Flash: 42.8 (#27), Qwen3.5 27B: 39.4 (#59)
| Benchmark | GLM-5.3-Flash | Qwen3.5 27B |
|---|---|---|
| LMArena Vision | 1296 | 1241 |
Multilingual GLM-5.3-Flash leads
GLM-5.3-Flash: 56.0 (#25), Qwen3.5 27B: 50.8 (#115)
| Benchmark | GLM-5.3-Flash | Qwen3.5 27B |
|---|---|---|
| LMArena Non-English | 1462 | 1390 |
| LMArena Chinese | 1527 | 1478 |
| LMArena French | 1496 | 1410 |
| LMArena German | 1470 | 1393 |
| LMArena Japanese | 1429 | 1345 |
| LMArena Korean | 1446 | 1358 |
| LMArena Russian | 1469 | 1390 |
| LMArena Spanish | 1471 | 1407 |
Instruction Following GLM-5.3-Flash leads
GLM-5.3-Flash: 77.5 (#20), Qwen3.5 27B: 73.5 (#119)
| Benchmark | GLM-5.3-Flash | Qwen3.5 27B |
|---|---|---|
| LMArena Instruction Following | 1478 | 1393 |
Long Context GLM-5.3-Flash leads
GLM-5.3-Flash: 45.4 (#39), Qwen3.5 27B: 43.1 (#106)
| Benchmark | GLM-5.3-Flash | Qwen3.5 27B |
|---|---|---|
| LMArena Longer Query | 1482 | 1413 |
Writing & Preference GLM-5.3-Flash leads
GLM-5.3-Flash: 65.3 (#50), Qwen3.5 27B: 59.3 (#111)
| Benchmark | GLM-5.3-Flash | Qwen3.5 27B |
|---|---|---|
| LMArena Text | 1471 | 1409 |
| LMArena Creative Writing | 1442 | 1362 |
| LMArena Multi-Turn | 1467 | 1410 |
Frequently asked questions
Is GLM-5.3-Flash better than Qwen3.5 27B?
GLM-5.3-Flash is the stronger model overall, scoring 51.8 to 41.9 on the Noometry Index.
Which is cheaper, GLM-5.3-Flash or Qwen3.5 27B?
GLM-5.3-Flash is cheaper. It lists at $0.15 per million input tokens and $0.50 per million output tokens; Qwen3.5 27B lists at $0.30 and $2.40.
Is GLM-5.3-Flash or Qwen3.5 27B better for coding?
GLM-5.3-Flash scores higher on coding benchmarks: 53.1 versus 38.9 in the Noometry coding category.
Which has the bigger context window?
GLM-5.3-Flash does, with 1M tokens against 262K.
How many benchmarks do GLM-5.3-Flash and Qwen3.5 27B share?
20 benchmarks have published results for both models. GLM-5.3-Flash has 40 scored results on Noometry and Qwen3.5 27B has 28.