Unverified50% confidenceBenchmarkExact time
在DeepSWE v1.1基准测试上，3.8 Flash超越了大多数体量更大的前沿模型
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/12/2026
First Seen
Valid until: 12/11/2026
Sources
Related Entities
Related Claims
UnverifiedV4.1 Flash在DeepSWE基准上以0.2分的微弱优势超过Opus 580% similarUnverifiedDeepSeek V4 Flash正式版在该测试中的速度比Pro预览版提升约3.6倍79% similarUnverifiedDeepSeek V4.1 Flash号称在能力、速度、成本三个维度上实现突破78% similarUnverified盘古2.0 Flash在GPQA Diamond科学推理测试中（Thinking模式）得分83.7，低于千问3.6 35B的86、27B稠密模型的87.8和DeepSeek V4 Flash的88.174% similarUnverifiedDeepSeek V4.1 Flash 在多项关键基准测试上能够与 GLM 5.3 和 Kimi K3 平起平坐73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/912483API
curl https://kongchang.com/api/v1/knowledge/claims/912483MCP
get_claim(id=912483)