Unverified40% confidenceFactTime unknown
Claude Opus 4.7在SWE Bench Pro上得分从53.4%提升至54.3%
1
Sources
40%
Confidence
Medium-term (~90 days)
Relevance
5/31/2026
First Seen
Valid until: 8/29/2026
Sources
Claude Opus 4.7深度实测：编码能力飙升，最强模型Mythos仍被封印
bilibili枫叶边城
Related Entities
Related Claims
UnverifiedClaude Opus 4.7在Cursor Bench上得分从58分飙升至70分84% similarUnverifiedClaude Opus 4.6在SWE-Bench Pro测试中得分为53.4%80% similarUnverifiedClaude Opus 4.8 scored 69.2% on SWEBench Pro, up from Opus 4.7's 64.3%80% similarUnverified据非官方爆料，Terminal Bench 3.0基准测试中Claude Opus 5 Max以43.5%的得分超越GPT-5.6-Sol的34.6%重登榜首74% similarUnverifiedClaude Opus 4.8 reached 85.9% on the GraphWalks 256K subset (up from 76.9% for 4.7) and 68.1% on the full 1-million-token version, nearly double Opus 4.7's score of 40.3%74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/13835API
curl https://kongchang.com/api/v1/knowledge/claims/13835MCP
get_claim(id=13835)