Unverified50% confidenceBenchmarkExact time
Grok 4.5在SWE Bench Pro任务上的输出Token比Opus 4.8少4.2倍
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/11/2026
First Seen
Valid until: 10/9/2026
Sources
Grok 4.5编码实测：性能、速度与成本全面解析
bilibiliAI_Express7/10/2026
Related Claims
UnverifiedGrok 4.5在Terminal Bench 2.1上取得83.3%，超越Opus 4.881% similarUnverifiedGrok 4.5在Deep SWE-Bench 1.0上取得62.0%，明显领先Opus 4.877% similarUnverified据称 Grok 4.5 的 token 效率约为同类领先模型的两倍，能用不到一半的步骤完成任务76% similarUnverifiedGrok 4.6的智能指数比Grok 4.5高5分，比Grok 4.3高23分69% similarUnverified马斯克发推透露Grok 4.7显著优于4.6，预计三到四周内就绪，并将加入来自SpaceX等旗下公司的补充训练数据66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/488058API
curl https://kongchang.com/api/v1/knowledge/claims/488058MCP
get_claim(id=488058)