# GLM-4.6 vs Phi 3 Mini 4k Instruct

> GLM-4.6 is the stronger model overall, scoring 41.4 to 27.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/glm-4-6-vs-phi-3-mini-4k-instruct
- Last updated: 2026-10-12
- Shared benchmarks: 17

## Summary

- They share 17 benchmarks with published results for both. GLM-4.6 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-4.6 leads 61.1 to 27.6.

## Snapshot

| | GLM-4.6 | Phi 3 Mini 4k Instruct |
|---|---|---|
| Provider | Z.ai (Zhipu) | Microsoft |
| Noometry Index | 41.4 | 27.9 |
| Rank | 135 | 328 |
| Context | 205K | — |
| Input $/M | $0.60 | — |
| Output $/M | $2.20 | — |
| Weights | Open | Open |

## Coding

- GLM-4.6: 40.1 (#148)
- Phi 3 Mini 4k Instruct: 26.6 (#323)

| Benchmark | GLM-4.6 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Coding | 1449 | 1093 |
| SWE-bench Verified (bash only) | 55.4% | — |
| LMArena WebDev | 1340 | — |
| SciCode | 38.4% | — |
| LiveBench Coding | — | 15.5% |
| ALE-Bench | 340.82 | — |
| HumanEval+ | — | 59.1% |
| MBPP+ | — | 54.2% |

## Agentic & Tool Use

- GLM-4.6: 32.3 (#66)
- Phi 3 Mini 4k Instruct: —

| Benchmark | GLM-4.6 | Phi 3 Mini 4k Instruct |
|---|---|---|
| Terminal-Bench | 24.5% | — |
| Berkeley Function Calling Leaderboard | 72.4% | — |

## Reasoning

- GLM-4.6: 23.7 (#172)
- Phi 3 Mini 4k Instruct: 14.1 (#328)

| Benchmark | GLM-4.6 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Hard Prompts | 1440 | 1072 |
| Kagi LLM Benchmark | 47.4% | — |
| CritPt | 1.1% | — |
| Chess Puzzles | — | 0% |
| LiveBench Reasoning | — | 26.8% |
| LiveBench Data Analysis | — | 34.7% |
| Adversarial NLI | — | 52.8% |
| BIG-Bench Hard | — | 71.7% |
| HellaSwag | — | 76.7% |
| LiveBench | — | 22.4% |
| WinoGrande | — | 70.8% |

## Math

- GLM-4.6: 39.1 (#111)
- Phi 3 Mini 4k Instruct: 26.6 (#257)

| Benchmark | GLM-4.6 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Math | 1432 | 1111 |
| LiveBench Math | — | 15.7% |
| FrontierMath (Feb 2025 set) | 3.8% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |

## Knowledge

- GLM-4.6: 40.2 (#124)
- Phi 3 Mini 4k Instruct: 28.5 (#246)

| Benchmark | GLM-4.6 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Expert | 1431 | 1045 |
| Vectara Hallucination Rate | 9.5% | — |
| ARC (AI2) Challenge | — | 84.9% |
| MMLU | — | 68.8% |
| OpenBookQA | — | 88% |
| TriviaQA | — | 64% |

## Multilingual

- GLM-4.6: 53.5 (#66)
- Phi 3 Mini 4k Instruct: 26.3 (#280)

| Benchmark | GLM-4.6 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Non-English | 1426 | 1021 |
| LMArena Chinese | 1499 | 1021 |
| LMArena French | 1459 | 1076 |
| LMArena German | 1447 | 1044 |
| LMArena Japanese | 1393 | 935 |
| LMArena Korean | 1400 | 905 |
| LMArena Russian | 1419 | 1022 |
| LMArena Spanish | 1436 | 1085 |

## Instruction Following

- GLM-4.6: 74.3 (#98)
- Phi 3 Mini 4k Instruct: 47.7 (#303)

| Benchmark | GLM-4.6 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Instruction Following | 1410 | 1053 |
| LiveBench Instruction Following | — | 39.1% |

## Long Context

- GLM-4.6: 43.4 (#94)
- Phi 3 Mini 4k Instruct: 31.7 (#276)

| Benchmark | GLM-4.6 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Longer Query | 1422 | 1044 |

## Writing & Preference

- GLM-4.6: 61.1 (#90)
- Phi 3 Mini 4k Instruct: 27.6 (#300)

| Benchmark | GLM-4.6 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Text | 1440 | 1073 |
| LMArena Creative Writing | 1411 | 1037 |
| LMArena Multi-Turn | 1427 | 1018 |
| EQ-Bench Creative Writing | 1411 | — |
| LiveBench Language | — | 9.2% |

## FAQ

### Is GLM-4.6 better than Phi 3 Mini 4k Instruct?

GLM-4.6 is the stronger model overall, scoring 41.4 to 27.9 on the Noometry Index.

### Is GLM-4.6 or Phi 3 Mini 4k Instruct better for coding?

GLM-4.6 scores higher on coding benchmarks: 40.1 versus 26.6 in the Noometry coding category.

### How many benchmarks do GLM-4.6 and Phi 3 Mini 4k Instruct share?

17 benchmarks have published results for both models. GLM-4.6 has 29 scored results on Noometry and Phi 3 Mini 4k Instruct has 35.