GLM 5.1 vs Gemma 4 31B
Côte à côte. Chaque métrique. Chaque benchmark.
| Type | GLM 5.1 | Gemma 4 31B |
|---|---|---|
| Provider |  z-ai |  Google DeepMind |
| score moyen | 70.2 | 61.6 |
| Prix d'entrée | $0.95 | $0.13 |
| Prix de sortie | $3.15 | $0.38 |
| Fenêtre de contexte | 203K tokens (~101 books) | 262K tokens (~131 books) |
| Sorti le | 2026-04-07 | 2026-04-02 |
| Code source ouvert | Open Source | Open Source |
Scores de benchmark
8 benchmarks · GLM 5.1: 8, Gemma 4 31B: 0
| Benchmark | Catégorie | GLM 5.1 | Gemma 4 31B |
|---|---|---|---|
| LiveBench — Agentic Coding | coding | 55.0 | 40.0 |
| LiveBench — Coding | coding | 75.4 | 60.3 |
| LiveBench — Data Analysis | reasoning | 63.2 | 58.8 |
| LiveBench — If | language | 68.5 | 67.6 |
| LiveBench — Language | language | 71.8 | 71.3 |
| LiveBench — Mathematics | math | 84.9 | 73.9 |
| LiveBench — Overall | knowledge | 70.2 | 61.6 |
| LiveBench — Reasoning | reasoning | 72.5 | 59.4 |