WHAT NEW IN X11.01 SUN

AI collections from X · 22 collections · 64,071 posts

实测IMAGE

DeepSeek V4.1 Flash fixes 42.7 of 105 bugs, edging Opus 5.5 on Bug Hunt

On 105 planted bugs across two production repos, DeepSeek V4.1 Flash 1017 fixed 42.7, slightly above Opus 5.5's 41.7 and Kimi K3.1 Pro's 36, at roughly $1–2 cost.

SuSu_酥酥👅
@NFT_ChenArchived 7 days agoOriginally posted 2026.10.24
DeepSeek V4.1 Flash fixes 42.7 of 105 bugs, edging Opus 5.5 on Bug Hunt
SuSu_酥酥👅 @NFT_ChenOriginally posted 2026.10.24
Open original on X

😧惊了!实际工作实测打脸:DeepSeek V4.1 Flash 1017 仍大幅优于 Kimi K3.1 Pro! 独立测试者 @PawelHuryn 在真实代码库上跑了 Bug Hunt 基准(2个生产级仓库 + 105个此前前沿模型漏掉的植入bug,要求找并修复): 1️⃣DeepSeek V4.1 Flash 1017 (max):修复 42.7/105(DeepSeek harness) 2️⃣Muse Wonder Full 1.0 (max):42 3️⃣Opus 5.5 (max):41.7 4️⃣Kimi K3.1 Pro (max):仅 36/105 5️⃣Muse Spark 1.3 (max):32.2 成本更夸张:DeepSeek 这次跑下来仅约 $1-2,Kimi 约 $14,其他模型动辄几十美元。 测试者直接说:“考虑成本,我看不出用这个模型的理由。” 实际修bug、读真实代码库场景,DeepSeek 依然性价比碾压。Kimi 还没打到这个点。 完整榜单:https://bughunt.productcompass.pm/ #DeepSeek #Kimi #AICoding #LLM #BugHunt #开源AI #DeepSeekV4 #梁文峰 #杨植麟

This post quotes another post: “DeepSeek V4.1 Flash tops Opus 5.5 on a 105-bug repair test”

x.com/nft_chen/status/2113823686130278548Archived 2026.10.24 12:05