Unverified60% confidenceFactExact time
The tester scored Claude Opus 4.8's one-shot tower defense game generation at 80 out of 100.
2
Sources
60%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026
Sources
Related Claims
UnverifiedClaude Opus 4.8在塔防游戏开发测试中获得80分评价,能生成核心功能和音效但炮台无法发射炮弹78% similarUnverified在SABench基准测试中,Claude Opus 4.7得分约33分,而人类高级工程师得分80-90分70% similarUnverifiedClaude Opus 4.8在智能体实战跑分中拿下1890分,比上一代高出137分,甩开GPT 5.5达121分68% similarPartially VerifiedIn a one-shot tower defense game generation test, Claude Opus 4.8 correctly handled tower placement, mirror zone setup, and different turret selection, and auto-generated sound effects.65% similarVerifiedClaude 4.5 Opus 在 SWE-Bench Verified 上达到了 80.9% 的成绩64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/59395API
curl https://kongchang.com/api/v1/knowledge/claims/59395MCP
get_claim(id=59395)