GLM 5.1
Código abertopor z-ai · Lançado em 2026-04-07
70.2
pontuação média
$0.95/1M
Preço de entrada
$3.15/1M
Preço de saída
203K tokens (~101 books)
Janela de contexto
text
Tipo
Tested on 12 benchmarks with 70.2% average. Top scores: Chatbot Arena Elo — Overall (1467.4%), LiveBench — Mathematics (84.9%), LiveBench — Coding (75.4%).
Pontuações de benchmark
| Benchmark | Categoria | Pontuação | Bar |
|---|---|---|---|
| Chatbot Arena Elo — Overall | arena | 1467.4 | |
| LiveBench — Mathematics | math | 84.9 | |
| LiveBench — Coding | coding | 75.4 | |
| LiveBench — Reasoning | reasoning | 72.5 | |
| LiveBench — Language | language | 71.8 | |
| LiveBench — Overall | knowledge | 70.2 | |
| LiveBench — If | language | 68.5 | |
| Artificial Analysis — Agentic Index | speed | 67.0 | |
| LiveBench — Data Analysis | reasoning | 63.2 | |
| LiveBench — Agentic Coding | coding | 55.0 | |
| Artificial Analysis — Quality Index | speed | 51.4 | |
| Artificial Analysis — Coding Index | speed | 43.4 |
