# GLM-4.5 vs Phi-4

> GLM-4.5 is the stronger model overall, scoring 42.0 to 31.2 on the Noometry Index. Phi-4 costs 11× less per token, which makes it the better buy when GLM-4.5's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/glm-4-5-vs-phi-4
- Last updated: 2026-10-11
- Shared benchmarks: 19

## Summary

- They share 19 benchmarks with published results for both. GLM-4.5 scores higher in 8 categories and Phi-4 in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GLM-4.5 leads 39.0 to 20.8.
- The biggest single-benchmark swing is Confabulations: 11.3% for GLM-4.5 and 29.4% for Phi-4.
- Phi-4 is cheaper at $0.07 / $0.14 per million input/output tokens, against $0.60 / $2.20 for GLM-4.5.
- GLM-4.5 accepts more context: 131K tokens versus 128K.

## Snapshot

| | GLM-4.5 | Phi-4 |
|---|---|---|
| Provider | Z.ai (Zhipu) | Microsoft |
| Noometry Index | 42.0 | 31.2 |
| Rank | 122 | 279 |
| Context | 131K | 128K |
| Input $/M | $0.60 | $0.07 |
| Output $/M | $2.20 | $0.14 |
| Weights | Open | Open |

## Coding

- GLM-4.5: 41.4 (#125)
- Phi-4: 34.4 (#239)

| Benchmark | GLM-4.5 | Phi-4 |
|---|---|---|
| LMArena Coding | 1434 | 1231 |
| SWE-bench Verified (bash only) | 54.2% | — |
| WeirdML | 40.6% | — |
| BigCodeBench Instruct | — | 45.5% |
| LiveBench Coding | — | 30.7% |
| BigCodeBench Complete | — | 55.4% |
| ALE-Bench | 344.82 | — |
| AlgoTune | 1.52 | — |

## Agentic & Tool Use

- GLM-4.5: —
- Phi-4: 22.8 (#128)

| Benchmark | GLM-4.5 | Phi-4 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 28.8% |
| BALROG | — | 11.6% |

## Reasoning

- GLM-4.5: 28.6 (#100)
- Phi-4: 17.7 (#291)

| Benchmark | GLM-4.5 | Phi-4 |
|---|---|---|
| LMArena Hard Prompts | 1429 | 1220 |
| Kagi LLM Benchmark | 57.9% | — |
| Chess Puzzles | — | 1% |
| LiveBench Reasoning | — | 47.8% |
| LiveBench Data Analysis | — | 45.2% |
| Epoch Capabilities Index | — | 130.42 |
| LiveBench | — | 41.6% |

## Math

- GLM-4.5: 39.0 (#116)
- Phi-4: 20.8 (#285)

| Benchmark | GLM-4.5 | Phi-4 |
|---|---|---|
| LMArena Math | 1427 | 1246 |
| OTIS Mock AIME 2024-2025 | — | 13.8% |
| LiveBench Math | — | 42% |
| MATH Level 5 | — | 64.9% |

## Knowledge

- GLM-4.5: 35.9 (#179)
- Phi-4: 32.6 (#209)

| Benchmark | GLM-4.5 | Phi-4 |
|---|---|---|
| Confabulations | 11.3% | 29.4% |
| LMArena Expert | 1433 | 1203 |
| GPQA Diamond | — | 56.1% |
| Humanity's Last Exam | 8.3% | — |
| Vectara Hallucination Rate | — | 3.7% |
| MMLU | — | 84.8% |

## Multilingual

- GLM-4.5: 52.8 (#77)
- Phi-4: 37.2 (#237)

| Benchmark | GLM-4.5 | Phi-4 |
|---|---|---|
| LMArena Non-English | 1417 | 1197 |
| LMArena Chinese | 1465 | 1212 |
| LMArena French | 1418 | 1224 |
| LMArena German | 1407 | 1222 |
| LMArena Japanese | 1415 | 1158 |
| LMArena Korean | 1380 | 1151 |
| LMArena Russian | 1414 | 1209 |
| LMArena Spanish | 1454 | 1234 |

## Instruction Following

- GLM-4.5: 74.1 (#104)
- Phi-4: 60.4 (#251)

| Benchmark | GLM-4.5 | Phi-4 |
|---|---|---|
| LMArena Instruction Following | 1404 | 1201 |
| LiveBench Instruction Following | — | 58.4% |

## Long Context

- GLM-4.5: 38.2 (#201)
- Phi-4: 36.9 (#226)

| Benchmark | GLM-4.5 | Phi-4 |
|---|---|---|
| LMArena Longer Query | 1412 | 1217 |
| Fiction.LiveBench | 58.3% | — |

## Writing & Preference

- GLM-4.5: 57.5 (#127)
- Phi-4: 40.5 (#244)

| Benchmark | GLM-4.5 | Phi-4 |
|---|---|---|
| LMArena Text | 1430 | 1217 |
| LMArena Creative Writing | 1395 | 1182 |
| Short-Story Creative Writing | 73.4% | 62.6% |
| LMArena Multi-Turn | 1415 | 1206 |
| EQ-Bench Creative Writing | 1343 | — |
| LiveBench Language | — | 25.6% |

## FAQ

### Is GLM-4.5 better than Phi-4?

GLM-4.5 is the stronger model overall, scoring 42.0 to 31.2 on the Noometry Index. Phi-4 costs 11× less per token, which makes it the better buy when GLM-4.5's lead doesn't matter for your workload.

### Which is cheaper, GLM-4.5 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 lists at $0.60 and $2.20.

### Is GLM-4.5 or Phi-4 better for coding?

GLM-4.5 scores higher on coding benchmarks: 41.4 versus 34.4 in the Noometry coding category.

### Which has the bigger context window?

GLM-4.5 does, with 131K tokens against 128K.

### How many benchmarks do GLM-4.5 and Phi-4 share?

19 benchmarks have published results for both models. GLM-4.5 has 27 scored results on Noometry and Phi-4 has 37.