Model comparison
GLM-4.5-Air vs Qwen3 14B
GLM-4.5-Air is the stronger model overall, scoring 38.9 to 35.5 on the Noometry Index.
Last verified . 2 shared benchmarks.
Alibaba (Qwen)Summary
- They share 2 benchmarks with published results for both. GLM-4.5-Air scores higher in 2 categories and Qwen3 14B in 3 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GLM-4.5-Air leads 24.1 to 18.5.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 43% for GLM-4.5-Air and 49.1% for Qwen3 14B.
- GLM-4.5-Air is cheaper at $0.20 / $1.10 per million input/output tokens, against $0.35 / $1.40 for Qwen3 14B.
Side by side
| GLM-4.5-Air | Qwen3 14B | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Alibaba (Qwen) |
| Noometry Index | 38.9 | 35.5 |
| Released | 2025-07-20 | 2025-04 |
| Weights | Open | Open |
| Context window | 131K | 131K |
| Max output | 98K | 8K |
| Input $ / M tokens | $0.20 | $0.35 |
| Output $ / M tokens | $1.10 | $1.40 |
| Results tracked | 27 | 12 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3 14B leads
GLM-4.5-Air: 33.3 (#259), Qwen3 14B: 37.3 (#195)
| Benchmark | GLM-4.5-Air | Qwen3 14B |
|---|---|---|
| SciCode | — | 31.6% |
| GSO | 2.9% | — |
| LMArena Coding | 1397 | — |
Agentic & Tool Use Not comparable
GLM-4.5-Air: —, Qwen3 14B: 29.6 (#83)
| Benchmark | GLM-4.5-Air | Qwen3 14B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 41% |
Reasoning GLM-4.5-Air leads
GLM-4.5-Air: 24.1 (#166), Qwen3 14B: 18.5 (#280)
| Benchmark | GLM-4.5-Air | Qwen3 14B |
|---|---|---|
| Kagi LLM Benchmark | 43% | 49.1% |
| CritPt | — | 0% |
| Chess Puzzles | — | 4% |
| LMArena Hard Prompts | 1379 | — |
| DTBench | — | 64% |
| LMCA | — | 18.2% |
| Epoch Capabilities Index | — | 138.23 |
| ForecastBench | 59.2 | — |
Math Qwen3 14B leads
GLM-4.5-Air: 36.2 (#170), Qwen3 14B: 38.6 (#133)
| Benchmark | GLM-4.5-Air | Qwen3 14B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 66.4% |
| Omni-MATH | 39.1% | — |
| LMArena Math | 1396 | — |
Knowledge Qwen3 14B leads
GLM-4.5-Air: 35.0 (#191), Qwen3 14B: 39.3 (#134)
| Benchmark | GLM-4.5-Air | Qwen3 14B |
|---|---|---|
| Vectara Hallucination Rate | 9.3% | 5.4% |
| GPQA Diamond | — | 63.8% |
| Humanity's Last Exam | 8.1% | — |
| MMLU-Pro | 76.2% | — |
| GPQA (HELM) | 59.4% | — |
| LMArena Expert | 1370 | — |
Multilingual Not comparable
GLM-4.5-Air: 49.1 (#135), Qwen3 14B: —
| Benchmark | GLM-4.5-Air | Qwen3 14B |
|---|---|---|
| LMArena Non-English | 1366 | — |
| LMArena Chinese | 1426 | — |
| LMArena French | 1399 | — |
| LMArena German | 1377 | — |
| LMArena Japanese | 1348 | — |
| LMArena Korean | 1308 | — |
| LMArena Russian | 1373 | — |
| LMArena Spanish | 1386 | — |
Instruction Following Not comparable
GLM-4.5-Air: 69.6 (#171), Qwen3 14B: —
| Benchmark | GLM-4.5-Air | Qwen3 14B |
|---|---|---|
| IFEval | 81.2% | — |
| LMArena Instruction Following | 1354 | — |
Long Context GLM-4.5-Air leads
GLM-4.5-Air: 41.6 (#135), Qwen3 14B: 38.1 (#204)
| Benchmark | GLM-4.5-Air | Qwen3 14B |
|---|---|---|
| Fiction.LiveBench | — | 62.5% |
| LMArena Longer Query | 1366 | — |
Writing & Preference Not comparable
GLM-4.5-Air: 55.9 (#139), Qwen3 14B: —
| Benchmark | GLM-4.5-Air | Qwen3 14B |
|---|---|---|
| LMArena Text | 1384 | — |
| LMArena Creative Writing | 1343 | — |
| WildBench | 78.9% | — |
| LMArena Multi-Turn | 1371 | — |
Frequently asked questions
Is GLM-4.5-Air better than Qwen3 14B?
GLM-4.5-Air is the stronger model overall, scoring 38.9 to 35.5 on the Noometry Index.
Which is cheaper, GLM-4.5-Air or Qwen3 14B?
GLM-4.5-Air is cheaper. It lists at $0.20 per million input tokens and $1.10 per million output tokens; Qwen3 14B lists at $0.35 and $1.40.
Is GLM-4.5-Air or Qwen3 14B better for coding?
Qwen3 14B scores higher on coding benchmarks: 37.3 versus 33.3 in the Noometry coding category.
Which has the bigger context window?
Both accept 131K tokens.
How many benchmarks do GLM-4.5-Air and Qwen3 14B share?
2 benchmarks have published results for both models. GLM-4.5-Air has 27 scored results on Noometry and Qwen3 14B has 12.