Model comparison
GLM-4.5-Air vs Phi-4
GLM-4.5-Air is the stronger model overall, scoring 38.9 to 31.2 on the Noometry Index. Phi-4 costs 4.9× less per token, which makes it the better buy when GLM-4.5-Air's lead doesn't matter for your workload.
Last verified . 18 shared benchmarks.
MicrosoftSummary
- They share 18 benchmarks with published results for both. GLM-4.5-Air scores higher in 7 categories and Phi-4 in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GLM-4.5-Air leads 36.2 to 20.8.
- The biggest single-benchmark swing is Vectara Hallucination Rate: 9.3% for GLM-4.5-Air and 3.7% for Phi-4.
- Phi-4 is cheaper at $0.07 / $0.14 per million input/output tokens, against $0.20 / $1.10 for GLM-4.5-Air.
- GLM-4.5-Air accepts more context: 131K tokens versus 128K.
Side by side
| GLM-4.5-Air | Phi-4 | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Microsoft |
| Noometry Index | 38.9 | 31.2 |
| Released | 2025-07-20 | 2024-12-11 |
| Weights | Open | Open |
| Context window | 131K | 128K |
| Max output | 98K | 4K |
| Input $ / M tokens | $0.20 | $0.07 |
| Output $ / M tokens | $1.10 | $0.14 |
| Results tracked | 27 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Phi-4 leads
GLM-4.5-Air: 33.3 (#259), Phi-4: 34.4 (#239)
| Benchmark | GLM-4.5-Air | Phi-4 |
|---|---|---|
| LMArena Coding | 1397 | 1231 |
| GSO | 2.9% | — |
| BigCodeBench Instruct | — | 45.5% |
| LiveBench Coding | — | 30.7% |
| BigCodeBench Complete | — | 55.4% |
Agentic & Tool Use Not comparable
GLM-4.5-Air: —, Phi-4: 22.8 (#128)
| Benchmark | GLM-4.5-Air | Phi-4 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 28.8% |
| BALROG | — | 11.6% |
Reasoning GLM-4.5-Air leads
GLM-4.5-Air: 24.1 (#166), Phi-4: 17.7 (#291)
| Benchmark | GLM-4.5-Air | Phi-4 |
|---|---|---|
| LMArena Hard Prompts | 1379 | 1220 |
| Kagi LLM Benchmark | 43% | — |
| Chess Puzzles | — | 1% |
| LiveBench Reasoning | — | 47.8% |
| LiveBench Data Analysis | — | 45.2% |
| Epoch Capabilities Index | — | 130.42 |
| ForecastBench | 59.2 | — |
| LiveBench | — | 41.6% |
Math GLM-4.5-Air leads
GLM-4.5-Air: 36.2 (#170), Phi-4: 20.8 (#285)
| Benchmark | GLM-4.5-Air | Phi-4 |
|---|---|---|
| LMArena Math | 1396 | 1246 |
| OTIS Mock AIME 2024-2025 | — | 13.8% |
| Omni-MATH | 39.1% | — |
| LiveBench Math | — | 42% |
| MATH Level 5 | — | 64.9% |
Knowledge GLM-4.5-Air leads
GLM-4.5-Air: 35.0 (#191), Phi-4: 32.6 (#209)
| Benchmark | GLM-4.5-Air | Phi-4 |
|---|---|---|
| Vectara Hallucination Rate | 9.3% | 3.7% |
| LMArena Expert | 1370 | 1203 |
| GPQA Diamond | — | 56.1% |
| Humanity's Last Exam | 8.1% | — |
| MMLU-Pro | 76.2% | — |
| Confabulations | — | 29.4% |
| GPQA (HELM) | 59.4% | — |
| MMLU | — | 84.8% |
Multilingual GLM-4.5-Air leads
GLM-4.5-Air: 49.1 (#135), Phi-4: 37.2 (#237)
| Benchmark | GLM-4.5-Air | Phi-4 |
|---|---|---|
| LMArena Non-English | 1366 | 1197 |
| LMArena Chinese | 1426 | 1212 |
| LMArena French | 1399 | 1224 |
| LMArena German | 1377 | 1222 |
| LMArena Japanese | 1348 | 1158 |
| LMArena Korean | 1308 | 1151 |
| LMArena Russian | 1373 | 1209 |
| LMArena Spanish | 1386 | 1234 |
Instruction Following GLM-4.5-Air leads
GLM-4.5-Air: 69.6 (#171), Phi-4: 60.4 (#251)
| Benchmark | GLM-4.5-Air | Phi-4 |
|---|---|---|
| LMArena Instruction Following | 1354 | 1201 |
| LiveBench Instruction Following | — | 58.4% |
| IFEval | 81.2% | — |
Long Context GLM-4.5-Air leads
GLM-4.5-Air: 41.6 (#135), Phi-4: 36.9 (#226)
| Benchmark | GLM-4.5-Air | Phi-4 |
|---|---|---|
| LMArena Longer Query | 1366 | 1217 |
Writing & Preference GLM-4.5-Air leads
GLM-4.5-Air: 55.9 (#139), Phi-4: 40.5 (#244)
| Benchmark | GLM-4.5-Air | Phi-4 |
|---|---|---|
| LMArena Text | 1384 | 1217 |
| LMArena Creative Writing | 1343 | 1182 |
| LMArena Multi-Turn | 1371 | 1206 |
| Short-Story Creative Writing | — | 62.6% |
| WildBench | 78.9% | — |
| LiveBench Language | — | 25.6% |
Frequently asked questions
Is GLM-4.5-Air better than Phi-4?
GLM-4.5-Air is the stronger model overall, scoring 38.9 to 31.2 on the Noometry Index. Phi-4 costs 4.9× less per token, which makes it the better buy when GLM-4.5-Air's lead doesn't matter for your workload.
Which is cheaper, GLM-4.5-Air or Phi-4?
Phi-4 is cheaper. It lists at $0.07 per million input tokens and $0.14 per million output tokens; GLM-4.5-Air lists at $0.20 and $1.10.
Is GLM-4.5-Air or Phi-4 better for coding?
Phi-4 scores higher on coding benchmarks: 34.4 versus 33.3 in the Noometry coding category.
Which has the bigger context window?
GLM-4.5-Air does, with 131K tokens against 128K.
How many benchmarks do GLM-4.5-Air and Phi-4 share?
18 benchmarks have published results for both models. GLM-4.5-Air has 27 scored results on Noometry and Phi-4 has 37.