Unverified85% confidenceFactTime unknown

M2.7在SWE-bench Pro基准测试中得分56.22%，超过了Gemini 3.1 Pro

1

Sources

85%

Confidence

Medium-term (~90 days)

Relevance

6/1/2026

First Seen

Valid until: 8/30/2026

Sources

MiniMax M2.7实测：编程能力比肩Claude Opus，成本低50倍

bilibiliAI-seeker

Related Entities

M2.7model SWE-bench Proconcept Gemini 3.1 Promodel

Related Claims

UnverifiedGemini 3.5 Pro在MMLU Pro基准测试中得分89.4分，领先GPT 5.5和Claude79% similar UnverifiedMiniMax M2.7在SWE-Pro基准测试上得分56.22%78% similar VerifiedGemini 3.1 Pro在高级推理基准测试中达到77%，而Claude Opus 4.6为68%77% similar UnverifiedMiniMax M2.7在VibePro基准测试上得分55.6%72% similar

Cite This Claim

Stable URI

https://kongchang.com/claim/21091

API

curl https://kongchang.com/api/v1/knowledge/claims/21091

MCP

get_claim(id=21091)