T
traeai
Sign in

模型

GPT-5.4

别名:GPT5.4

OpenAI开发的语言模型。

相关材料

已收录 8 条与 GPT-5.4 相关的内容,按评分排序。

科技爱好者周刊(第 397 期):财富正在向 AI 集中

科技爱好者周刊(第 397 期):财富正在向 AI 集中

阮一峰的网络日志4040 字 (约 17 分钟)
85

本期科技爱好者周刊聚焦于AI技术的快速发展及其对社会财富分配的影响。文章指出,AI相关产业如内存、储存、CPU、服务器等的股价大幅上涨,表明财富正迅速向AI领域集中。此外,文章还探讨了AI在日常生活中的应用,如通过AI估算食物的碳水含量,但实验表明AI在这方面并不准确。微软宣布将淘汰短信验证码,转而采用更安全的验证方式,如Passkey。亚马逊推出供应链服务,开放其物流网络,可能对制造业产生影响。最后,文章介绍了一个机械打字机模型玩具,以及几篇技术文章,包括关于GitHub Pages域名盗用问题、JavaScript ShadowRealm API和Firefox配置指南。

入选理由:AI相关产业的股价大幅上涨,表明财富正迅速向AI领域集中。

FeaturedArticle#AI#财富分配#微软#亚马逊#供应链服务#Passkey#GitHub Pages#ShadowRealm API#Firefox配置中文
Artificial Analysis放榜:千问3.7问鼎国产模型冠军,全球前五

Alibaba Cloud's Qwen3.7-Max scores 56.6 to rank 5th globally and 1st domestically in Artificial Analysis benchmark, soon available via Alibaba Cloud's BaiLian API.

入选理由:Qwen3.7-Max得分56.6分,超越国产所有模型,逼近GPT-5.4、Gemini3.1 Pro等国际顶尖模型

FeaturedArticle#Qwen3.7-Max#Artificial Analysis#Model Benchmarking#Alibaba Cloud中文
Parloa builds service agents customers want to talk to

Parloa builds service agents customers want to talk to

OpenAI Blog1594 字 (约 7 分钟)
85

Parloa’s AI Agent Management Platform (AMP) built on GPT-5.4 helps enterprises design, deploy, and manage voice-driven customer service systems with improved automation efficiency.

入选理由:Parloa 使用自然语言配置代替传统代码编写,降低非技术人员构建 AI 客服的门槛。

FeaturedArticle#AI#Customer Service#OpenAI#GPT-5.4英文
Are AI Labs Racing Toward Self-Improvement?

Are AI Labs Racing Toward Self-Improvement?

Last Week in AI239 字 (约 1 分钟)
75

AI labs are explicitly pursuing automated AI research to achieve fully self-improving AI software loops, aiming to trigger a software-only singularity and intelligence explosion, but current model improvement mechanisms have become opaque, mixing architectural changes, scale expansion, and more training data.

入选理由:AI实验室明确将自动化AI研究作为核心战略目标

FeaturedVideo#AI Research#Model Optimization#Singularity Theory#Automation#OpenAI英文
We analyzed GPT 5.5 vs GPT 5.4 and found that costs increased between 49-92%.

The 2x price hike of ...

We analyzed GPT 5.5 vs GPT 5.4 and found that costs increased between 49-92%

OpenRouter(@OpenRouterAI)150 字 (约 1 分钟)
62

OpenRouter found that while GPT-5.5's price doubled compared to GPT-5.4, its cost increase was mitigated by generating 19-34% fewer completion tokens on long prompts, resulting in a net cost rise of 49-92%.

入选理由:GPT-5.5价格较GPT-5.4翻倍,但单位任务成本仅上升49%-92%。

FeaturedTweet#GPT-5.5#GPT-5.4#OpenRouter#token efficiency#API cost英文
Anthropic’s Billion-Dollar Decision to Hold Back

Anthropic’s Billion-Dollar Decision to Hold Back

Last Week in AI243 字 (约 1 分钟)
40

Anthropic公司选择不完全发布Mythos,这一决策背后是出于安全优先于利润的考量。Mythos在基准测试中与Opus 4.6相当,在工具使用上超越GPT 5.4,展示了其潜在的巨大价值。

入选理由:Anthropic未全面发布Mythos,重视安全而非立即盈利。

FeaturedVideo#Anthropic#Mythos#AI Ethics#GPT#Opus英文

跨材料问答 · GPT-5.4

回答基于:GPT-5.4 相关 8 条材料
    0 / 500

    AI may generate inaccurate information. Please verify important content.