Unverified95% confidenceFactTime unknown

Gemini 3 Pro在五轮测试中总共消耗了22.3万个Token，是四个模型中最多的

1

Sources

95%

Confidence

Medium-term (~90 days)

Relevance

6/1/2026

First Seen

Valid until: 8/30/2026

Sources

前端UI生成实测：GLM 4.7、Claude Opus、Gemini、Minimax谁最强？

bilibiliGoldenSpiderAI

Related Entities

Gemini 3 Promodel

Related Claims

UnverifiedGemini Nano通过从Gemini系列大模型蒸馏而来，参数量控制在10亿至30亿级别74% similar UnverifiedGemini 3.1 Pro High在SWE-bench测试中通过率为84.17%，排名第一，生成代码量为307,000行，Bug密度为614个/百万行代码，安全问题为210个/百万行代码74% similar UnverifiedGemini 3在SWE-bench基准测试中智能体场景下的得分远超同类模型71% similar

Cite This Claim

Stable URI

https://kongchang.com/claim/20501

API

curl https://kongchang.com/api/v1/knowledge/claims/20501

MCP

get_claim(id=20501)