# Gemma 4 31B IT vs GLM-5.3-Flash

> GLM-5.3-Flash is the stronger model overall, scoring 49.6 to 43.2 on the Noometry Index. Gemma 4 31B IT costs 1.6× less per token, which makes it the better buy when GLM-5.3-Flash's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gemma-4-31b-it-vs-glm-5-3-flash
- Last updated: 2026-10-10
- Shared benchmarks: 24

## Summary

- They share 24 benchmarks with published results for both. Gemma 4 31B IT scores higher in 0 categories and GLM-5.3-Flash in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GLM-5.3-Flash leads 58.4 to 37.9.
- The biggest single-benchmark swing is Surface Evolver Bench: 30.6% for Gemma 4 31B IT and 52.5% for GLM-5.3-Flash.
- Gemma 4 31B IT is cheaper at $0.09 / $0.34 per million input/output tokens, against $0.15 / $0.50 for GLM-5.3-Flash.
- GLM-5.3-Flash accepts more context: 1M tokens versus 262K.

## Snapshot

| | Gemma 4 31B IT | GLM-5.3-Flash |
|---|---|---|
| Provider | Google | Z.ai (Zhipu) |
| Noometry Index | 43.2 | 49.6 |
| Rank | 85 | 45 |
| Context | 262K | 1M |
| Input $/M | $0.09 | $0.15 |
| Output $/M | $0.34 | $0.50 |
| Weights | Open | Open |

## Coding

- Gemma 4 31B IT: 41.9 (#98)
- GLM-5.3-Flash: 53.7 (#26)

| Benchmark | Gemma 4 31B IT | GLM-5.3-Flash |
|---|---|---|
| LMArena WebDev | 1364 | 1614 |
| SciCode | 43.4% | 51.6% |
| LMArena Coding | 1459 | 1506 |
| ALE-Bench | 925.5 | 303.55 |
| DeepSWE | — | 63.4% |
| FrontierCode | — | 31.8% |
| WeirdML | 52.3% | — |

## Agentic & Tool Use

- Gemma 4 31B IT: —
- GLM-5.3-Flash: 34.3 (#41)

| Benchmark | Gemma 4 31B IT | GLM-5.3-Flash |
|---|---|---|
| APEX-Agents | — | 52.8% |
| GDP.pdf | — | 14% |

## Reasoning

- Gemma 4 31B IT: 26.4 (#119)
- GLM-5.3-Flash: 31.7 (#69)

| Benchmark | Gemma 4 31B IT | GLM-5.3-Flash |
|---|---|---|
| CritPt | 1.4% | 15.4% |
| Chess Puzzles | 5% | 14% |
| LMArena Hard Prompts | 1448 | 1490 |
| Surface Evolver Bench | 30.6% | 52.5% |
| Epoch Capabilities Index | 142.74 | 151.9 |
| Kagi LLM Benchmark | 63.5% | — |
| NYT Connections (extended) | 70.6% | — |
| Thematic Generalization | 53% | — |
| Mystery Game Puzzles | — | 8% |
| DTBench | 82.7% | — |
| LMCA | 39.3% | — |
| Bench to the Future 3 | — | 0.15 |

## Math

- Gemma 4 31B IT: 41.3 (#72)
- GLM-5.3-Flash: 53.1 (#38)

| Benchmark | Gemma 4 31B IT | GLM-5.3-Flash |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.3% | 93.9% |
| LMArena Math | 1464 | 1499 |
| FrontierMath (Tiers 1-3) | — | 55.8% |
| FrontierMath Tier 4 | — | 17.1% |
| ProofBench | — | 21% |

## Knowledge

- Gemma 4 31B IT: 37.9 (#141)
- GLM-5.3-Flash: 58.4 (#33)

| Benchmark | Gemma 4 31B IT | GLM-5.3-Flash |
|---|---|---|
| GPQA Diamond | 75.8% | 90.2% |
| LMArena Expert | 1465 | 1512 |
| SimpleQA Verified | 10.4% | — |
| Vectara Hallucination Rate | 7.4% | — |

## Multimodal

- Gemma 4 31B IT: 41.0 (#32)
- GLM-5.3-Flash: 42.6 (#25)

| Benchmark | Gemma 4 31B IT | GLM-5.3-Flash |
|---|---|---|
| LMArena Vision | 1278 | 1304 |
| LMArena Document | 1425 | — |

## Multilingual

- Gemma 4 31B IT: 54.0 (#52)
- GLM-5.3-Flash: 56.2 (#24)

| Benchmark | Gemma 4 31B IT | GLM-5.3-Flash |
|---|---|---|
| LMArena Non-English | 1431 | 1462 |
| LMArena Chinese | 1478 | 1529 |
| LMArena French | 1436 | 1495 |
| LMArena Russian | 1462 | 1472 |
| LMArena Spanish | 1443 | 1471 |
| LMArena German | — | 1457 |
| LMArena Japanese | — | 1420 |
| LMArena Korean | — | 1443 |

## Instruction Following

- Gemma 4 31B IT: 75.6 (#58)
- GLM-5.3-Flash: 77.8 (#17)

| Benchmark | Gemma 4 31B IT | GLM-5.3-Flash |
|---|---|---|
| LMArena Instruction Following | 1433 | 1480 |

## Long Context

- Gemma 4 31B IT: 44.2 (#67)
- GLM-5.3-Flash: 45.3 (#41)

| Benchmark | Gemma 4 31B IT | GLM-5.3-Flash |
|---|---|---|
| LMArena Longer Query | 1446 | 1479 |

## Writing & Preference

- Gemma 4 31B IT: 60.7 (#92)
- GLM-5.3-Flash: 65.4 (#49)

| Benchmark | Gemma 4 31B IT | GLM-5.3-Flash |
|---|---|---|
| LMArena Text | 1443 | 1471 |
| LMArena Creative Writing | 1416 | 1440 |
| LMArena Multi-Turn | 1452 | 1471 |
| EQ-Bench Creative Writing | 1368 | — |
| EQ-Bench 4 | 1120 | — |

## FAQ

### Is Gemma 4 31B IT better than GLM-5.3-Flash?

GLM-5.3-Flash is the stronger model overall, scoring 49.6 to 43.2 on the Noometry Index. Gemma 4 31B IT costs 1.6× less per token, which makes it the better buy when GLM-5.3-Flash's lead doesn't matter for your workload.

### Which is cheaper, Gemma 4 31B IT or GLM-5.3-Flash?

Gemma 4 31B IT is cheaper. It lists at $0.09 per million input tokens and $0.34 per million output tokens; GLM-5.3-Flash lists at $0.15 and $0.50.

### Is Gemma 4 31B IT or GLM-5.3-Flash better for coding?

GLM-5.3-Flash scores higher on coding benchmarks: 53.7 versus 41.9 in the Noometry coding category.

### Which has the bigger context window?

GLM-5.3-Flash does, with 1M tokens against 262K.

### How many benchmarks do Gemma 4 31B IT and GLM-5.3-Flash share?

24 benchmarks have published results for both models. Gemma 4 31B IT has 35 scored results on Noometry and GLM-5.3-Flash has 36.