GLM 4.6
开源来自 z-ai · 发布于 2025-09-30
50.8
平均分
$0.39/1M
输入价格
$1.90/1M
输出价格
205K tokens (~102 books)
上下文窗口
text
类型
Tested on 20 benchmarks with 50.8% average. Top scores: Chatbot Arena Elo — Overall (1425.8%), Chatbot Arena Elo — Coding (1353.7%), OpenCompass — AIME2025 (90.3%).
基准测试分数
| 基准测试 | 类别 | 分数 | Bar |
|---|---|---|---|
| Chatbot Arena Elo — Overall | arena | 1425.8 | |
| Chatbot Arena Elo — Coding | arena | 1353.7 | |
| OpenCompass — AIME2025 | math | 90.3 | |
| OpenCompass — IFEval | language | 88.7 | |
| OpenCompass — MMLU-Pro | knowledge | 83.0 | |
| LiveBench — Mathematics | math | 81.1 | |
| OpenCompass — GPQA-Diamond | knowledge | 80.4 | |
| OpenCompass — LiveCodeBenchV6 | coding | 78.2 | |
| LiveBench — Coding | coding | 71.0 | |
| LiveBench — Reasoning | reasoning | 62.1 | |
| LiveBench — Language | language | 59.0 | |
| LiveBench — Overall | knowledge | 55.2 | |
| LiveBench — Data Analysis | reasoning | 52.0 | |
| LiveBench — Agentic Coding | coding | 35.0 | |
| LiveBench — If | language | 26.2 | |
| Terminal Bench | coding | 24.5 | |
| OpenCompass — HLE | knowledge | 19.3 | |
| FrontierMath-2025-02-28-Private | math | 3.8 | |
| APEX-Agents | agentic | 3.0 | |
| FrontierMath-Tier-4-2025-07-01-Private | math | 2.1 |