# GLM-5 vs Phi 3 Mini 4k Instruct

> GLM-5 is the stronger model overall, scoring 45.9 to 27.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/glm-5-vs-phi-3-mini-4k-instruct
- Last updated: 2026-10-10
- Shared benchmarks: 18

## Summary

- They share 18 benchmarks with published results for both. GLM-5 scores higher in 8 categories and Phi 3 Mini 4k Instruct in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GLM-5 leads 66.2 to 27.7.
- The biggest single-benchmark swing is Chess Puzzles: 10% for GLM-5 and 0% for Phi 3 Mini 4k Instruct.

## Snapshot

| | GLM-5 | Phi 3 Mini 4k Instruct |
|---|---|---|
| Provider | Z.ai (Zhipu) | Microsoft |
| Noometry Index | 45.9 | 27.4 |
| Rank | 59 | 324 |
| Context | 205K | — |
| Input $/M | $1 | — |
| Output $/M | $3.20 | — |
| Weights | Open | Open |

## Coding

- GLM-5: 49.0 (#47)
- Phi 3 Mini 4k Instruct: 26.1 (#319)

| Benchmark | GLM-5 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Coding | 1461 | 1093 |
| SWE-bench Verified | 72.1% | — |
| SWE-bench Verified (bash only) | 72.8% | — |
| LMArena WebDev | 1434 | — |
| SWE-bench Multilingual | 69.7% | — |
| WeirdML | 48.2% | — |
| LiveBench Coding | — | 15.5% |
| ALE-Bench | 765.62 | — |
| HumanEval+ | — | 59.1% |
| MBPP+ | — | 54.2% |

## Agentic & Tool Use

- GLM-5: 30.9 (#65)
- Phi 3 Mini 4k Instruct: —

| Benchmark | GLM-5 | Phi 3 Mini 4k Instruct |
|---|---|---|
| Terminal-Bench | 52.4% | — |
| τ²-bench Airline | 82.5% | — |
| τ²-bench Banking | 9.8% | — |
| τ²-bench Retail | 73.7% | — |
| τ²-bench Telecom | 86.8% | — |
| Vending-Bench 2 | 4,432 | — |

## Reasoning

- GLM-5: 27.6 (#100)
- Phi 3 Mini 4k Instruct: 13.6 (#326)

| Benchmark | GLM-5 | Phi 3 Mini 4k Instruct |
|---|---|---|
| Chess Puzzles | 10% | 0% |
| LMArena Hard Prompts | 1451 | 1072 |
| ARC-AGI-2 | 4.9% | — |
| SimpleBench | 53.2% | — |
| Kagi LLM Benchmark | 75% | — |
| NYT Connections (extended) | 74.8% | — |
| ARC-AGI-1 | 44.7% | — |
| LiveBench Reasoning | — | 26.8% |
| LiveBench Data Analysis | — | 34.7% |
| Adversarial NLI | — | 52.8% |
| BIG-Bench Hard | — | 71.7% |
| Epoch Capabilities Index | 145.84 | — |
| ForecastBench | 61 | — |
| HellaSwag | — | 76.7% |
| LiveBench | — | 22.4% |
| WinoGrande | — | 70.8% |

## Math

- GLM-5: 44.8 (#66)
- Phi 3 Mini 4k Instruct: 24.5 (#258)

| Benchmark | GLM-5 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Math | 1440 | 1111 |
| MathArena Final-Answer Competitions | 65.7% | — |
| OTIS Mock AIME 2024-2025 | 80% | — |
| LiveBench Math | — | 15.7% |
| FrontierMath (Feb 2025 set) | 16.4% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |

## Knowledge

- GLM-5: 52.2 (#59)
- Phi 3 Mini 4k Instruct: 27.9 (#241)

| Benchmark | GLM-5 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Expert | 1454 | 1045 |
| GPQA Diamond | 87.8% | — |
| Vectara Hallucination Rate | 10.1% | — |
| ARC (AI2) Challenge | — | 84.9% |
| MMLU | — | 68.8% |
| OpenBookQA | — | 88% |
| TriviaQA | — | 64% |

## Multilingual

- GLM-5: 53.9 (#57)
- Phi 3 Mini 4k Instruct: 26.5 (#275)

| Benchmark | GLM-5 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Non-English | 1430 | 1021 |
| LMArena Chinese | 1511 | 1021 |
| LMArena French | 1455 | 1075 |
| LMArena German | 1445 | 1044 |
| LMArena Japanese | 1416 | 935 |
| LMArena Korean | 1421 | 905 |
| LMArena Russian | 1436 | 1022 |
| LMArena Spanish | 1452 | 1084 |

## Instruction Following

- GLM-5: 75.4 (#64)
- Phi 3 Mini 4k Instruct: 48.0 (#298)

| Benchmark | GLM-5 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Instruction Following | 1427 | 1053 |
| LiveBench Instruction Following | — | 39.1% |

## Long Context

- GLM-5: 44.6 (#58)
- Phi 3 Mini 4k Instruct: 31.7 (#271)

| Benchmark | GLM-5 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Longer Query | 1446 | 1044 |
| CL-bench | 18.7% | — |

## Writing & Preference

- GLM-5: 66.2 (#38)
- Phi 3 Mini 4k Instruct: 27.7 (#295)

| Benchmark | GLM-5 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Text | 1446 | 1073 |
| LMArena Creative Writing | 1438 | 1037 |
| LMArena Multi-Turn | 1456 | 1018 |
| EQ-Bench Creative Writing | 1601 | — |
| LiveBench Language | — | 9.2% |

## FAQ

### Is GLM-5 better than Phi 3 Mini 4k Instruct?

GLM-5 is the stronger model overall, scoring 45.9 to 27.4 on the Noometry Index.

### Is GLM-5 or Phi 3 Mini 4k Instruct better for coding?

GLM-5 scores higher on coding benchmarks: 49.0 versus 26.1 in the Noometry coding category.

### How many benchmarks do GLM-5 and Phi 3 Mini 4k Instruct share?

18 benchmarks have published results for both models. GLM-5 has 45 scored results on Noometry and Phi 3 Mini 4k Instruct has 35.