Unverified75% confidenceFactTime unknown
Gemma 4的31B模型在M2026评测中得分89.2%，代码模型的Elo评分约为2500
1
Sources
75%
Confidence
Medium-term (~90 days)
Relevance
6/3/2026
First Seen
Valid until: 9/1/2026
Sources
Ollama+Gemma 4本地运行Codex：零成本AI编程完整指南
bilibili薛定猫AI
Related Entities
Related Claims
UnverifiedGemma 4 31B 的综合能力评分达到 11,452 分，在同级别模型中处于领先地位76% similarUnverifiedGemma 2B模型拥有约20亿参数73% similarUnverifiedGemma 4的E2B版本在社区基准测试中多项任务表现堪比Gemma 3的271亿参数模型72% similarUnverifiedGemma 4 31B 在评测中超越了智谱 GLM、DeepSeek V3.2 等同参数量级模型，并与 Gemini 5、Kimi 2.5 等大模型表现相当70% similarUnverifiedGemma 4的31B Dense模型在开源模型排行榜上位列第三70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/38840API
curl https://kongchang.com/api/v1/knowledge/claims/38840MCP
get_claim(id=38840)