Unverified60% confidenceFactTime unknown
GLM 4.6在HumanEval、MBPP、SWE-bench等代码评测基准以及MMLU、GPQA等综合能力基准上的表现对齐Claude 3.4和Claude 3.4.5
2
Sources
60%
Confidence
Medium-term (~90 days)
Relevance
6/3/2026
First Seen
Valid until: 9/1/2026
Sources
Claude Code搭配GLM4.6：编程Agent国产化部署指南
bilibili潜龙悟用-
Related Entities
Related Claims
UnverifiedGLM-4.7在代码能力基准测试中超过DeepSeek V3.2及Claude Sonnet 4.577% similarUnverifiedGLM4在代码推理和通用知识测试中得分74.4%，仅次于Claude 3 Opus76% similarUnverified在MMLU、HumanEval等主流基准测试上，Claude 3.5、GPT-4o、Gemini 1.5 Pro、Grok-2之间的分差已压缩至统计误差范围内72% similarUnverifiedGLM-4.7是一款MOE（混合专家）架构模型，总参数量达358B（3580亿）72% similarUnverified智谱GLM系列从GLM4.5到4.7在358B参数规模上连续打磨三代，随后GLM5翻倍至754B71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/38492API
curl https://kongchang.com/api/v1/knowledge/claims/38492MCP
get_claim(id=38492)