大模型
x.com ·
@Arena.ai (@arena)
DeepSeek-V4.1-Flash 在 Code Arena WebDev 早期 AutoEval 中约列第 14
Arena 称 DeepSeek-V4.1-Flash 在 Code Arena: WebDev 的 AutoEval 早期评测中获 1620 分、总榜约第 14,开源模型中约第 4,距 Qwen3.8-Flash-Next 约 11 分;相比 V4-Flash(High)提升 38 分、相比 V4-Pro(High)提升 40 分,并与第 11–13 名相差 5 分以内,输入/输出价格为 $0.30/$1.20。Arena 提示该分数由基于人类偏好数据训练的 Reward Model 自动投票产生,后续将随真人投票变化。
AI推荐理由 Arena 官方账号给出 DeepSeek-V4.1-Flash 在 Code Arena: WebDev 的 AutoEval 早期得分 1620 分、约第 14 名,开源模型中约第 4,并列出与 V4-Flash、V4-Pro 的分数差及与第 11–13 名的价格对比。信息密度较高且含具体数字,但明确标注为 Reward Model 自动投票的早期分数,需等待真人投票收敛,因此结论具有不确定性。
74
浏览 103,531
点赞 832
评论 36
转发 57
正文
DeepSeek-V4.1-Flash by @deepseek_ai just landed ~#14 overall in Code Arena: WebDev with 1620 pts (AutoEval)! Among open models, DeepSeek-V4.1-Flash landed at ~#4 within 11 pts of Qwen3.8-Flash-Next.
This release is a significant improvement compared to DeepSeek-V4 variants:
+38 pts vs. V4-Flash (High) at #20 (1582 pts)
+40 pts vs. V4-Pro (High) #21 (1580 pts)
DeepSeek-V4.1-Flash is within 5 pts of the models ranked #11–#13, while costing substantially less (priced per million input/output tokens):
~#14 DeepSeek-V4.1-Flash: $0.30/ $1.20
#13 Hy4 Preview: $0.83/ $2.50
#12 Grok-4.6 High: $2/ $6
#11 Muse Spark 1.3 xHigh: $1.25/ $4.25
Note: this is an early AutoEval score, in which a Reward Model trained on Arena’s human preference data casts automatic votes in place of live votes. We’ll continue to see how scores converge as more live human votes come in.
Congrats to the @deepseek_ai team on this contribution to the open source ecosystem!