加载中...
加载中...
Google Gemma 4 31B
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
Gemma 4 31B currently shows benchmark results led by MMLU Pro (16 / 115, score 85.20), LiveCodeBench (21 / 108, score 80), GPQA Diamond (39 / 162, score 84.30). This page also consolidates core specs, context limits, and API pricing so you can evaluate the model from benchmark results and deployment constraints together.
2026年4月2日,Google DeepMind 携全系 Gemma 4 开源模型亮相。31B 是该系列的旗舰型号,采用了纯密集型网络(Dense)架构。它继承了与当前商业模型 Gemini 3 相同的先进底层研究成果,代表了当今开源生态中该尺寸级别内最强的前沿水平(Frontier AI)。
31B 专注于提供“开箱即用”的高保真知识与极深度的思维解析:
欢迎关注 DataLearner 官方微信,获得最新 AI 技术推送
