跳到正文
7月3日周五
7月2日周四
  1. Ethan Mollick72

    Ethan Mollick 提醒,Fable 在长时间任务中会发展出独有的内部节奏和对话模式,其多个 agent 与任务相互强化,使得 "Claudish" 语言越来越 "Claudish"。若不主动要求它用 plain language 报告,这种怪异风格会渗透到包括菜单在内的所有输出中。

    引用Ethan Mollick@emollick

    One thing I mentioned only in passing in my Fable post is that, for long running tasks, Fable starts to develop its own dialect as its many agents and tasks reinforce themselves and make Claudish language ever more Claudish. You need to ask it to report out in plain English.

  2. Greg Brockman47

    Codex 现在每天早上为我生成一份“日报”,包含未读消息、日历、冲浪报告和新闻。一切能让我直到当天晚些时候都不碰手机的事情都是优先事项。Greg Brockman 表示这是利用 Codex 制作个性化每日摘要的方法。

    引用Ryan Doyle@doooyle

    surprised more people aren't doing something like this Codex now creates a "newspaper" for me every morning Unread messages, calendar, surf report, news Anything I can do to stay off my phone until later in the day is a priority

  3. Rohan Paul53

    在 atomic.chat(本地 LLM 桌面应用)的 HTML5 物理竞赛中,Fable 5 以 A+ 成绩完成全部三个场景(火车脱轨、汽车空中碰撞、怪物卡车碾压),消耗 62,158 token,成本 $3.12。相比之下,Opus 4.8 消耗 22,280 token/$0.56,GPT 5.5 消耗 37,753 token/$1.14(在怪物卡车场景中略胜 Fable),GLM 5.2 消耗 36,246 token/$0.08 但未赢得任何场景。Fable 5 质量最佳但成本最高。

    引用atomic.chat@atomic_chat_hq

    Fable 5 totally crushed our new contest, but it cost 6x more than Opus 4.8! We gave 4 models the same prompt: build three self-contained HTML5 canvas scenes with real physics demos Prompts: — A train derailing off a broken bridge into the water — Two cars jumping off ramps and colliding mid-air over a canyon — A monster truck crushing a row of parked cars Outputs: Fable 5: 62,158 tokens, $3.12 GPT 5.5: 37,753 tokens, $1.14 Opus 4.8: 22,280 tokens, $0.56 GLM 5.2: 36,246 tokens, $0.08 Fable 5 did all three scenes at A+. The crashes looked real, things fell and broke the right way, and nothing went through the ground or floated. GPT 5.5 was the closest to Fable. In the Bigfoot show, we think GPT was even a little better. GLM 5.2 did not win any scene, but it was the cheapest by far. Fable is the best pick for quality, but you pay more for it.

7月1日周三
  1. OpenBMB35

    真的写得很好且实用。👍 我们很欣赏你如何拆解整个本地管道,并突出说明了为什么 MiniCPM-V 4.6 是边缘部署的绝佳选择——其小巧的尺寸、强大的多模态能力以及在 Apple Silicon 上出色的效率正是这些模型令人兴奋之处。🤗

    引用Shirish Srivastava@Shreesh87

    https://medium.com/@shirishsrivastava/running-local-vision-llms-on-apple-silicon-a-lightweight-playwright-llama-cpp-pipeline-f13265f3f06a @llama @MiniCPM

  2. Noah Zweben31

    Claude Tag 现允许用户在管理后台(claude.ai/admin-settings)为频道或工作区默认切换模型与努力级别。此前用户 @morganlinton 反馈该工具默认使用 Opus + High Effort 导致 token 消耗过高,而 Claude 自身推荐约 90% 任务使用 Sonnet。新设置可直接在“Select scope > Advanced”中调整。

    引用Morgan@morganlinton

    I started using Claude Tag with our team in Slack, and everyone is really liking it so far. It's definitely a notch above Slack's built-in AI functionality. The one thing I can't quite figure out is how to change the model/effort level. I was able to run a bunch of tasks we are looking at doing with Claude Tag, by Claude itself, and it recommended Sonnet for about 90% of them. Right now it looks like Claude Tag is defaulting to Opus, High Effort. I do think if we did use Opus with it, most tasks wouldn't need High Effort, Medium would be more than enough. Does anyone know if there's a way to update the model/effort level? If we do everything we want to do with Claude Tag, on Opus, it's going to use a lot more tokens than we would need to use, and would run up a pretty hefty bill. Thanks in advance to anyone who can point me in the right direction!

  3. Rohan Paul58

    atomic.chat桌面应用对Claude Sonnet 5、Opus 4.8、Sonnet 4.6及GPT 5.5进行对比测试。使用同一提示词构建三个HTML5物理碰撞演示(汽车撞墙、破坏球毁屋、投石机砸城)。Sonnet 5在全部测试中与GPT 5.5和Opus 4.8表现相当,其中破坏球场景胜Opus 4.8,投石机场景胜GPT 5.5。Sonnet 5仅用15,047 tokens($0.15),GPT 5.5使用31,152 tokens($0.94),成本低约6倍;Opus 4.8使用23,063 tokens($0.58),Sonnet 4.6使用25,824 tokens($0.39)。Sonnet 5 token消耗最少,图形细节仍有提升空间。

    引用atomic.chat@atomic_chat_hq

    New Claude Sonnet 5 performs at GPT 5.5 level 6x cheaper! We gave 4 models the same prompt: build three self-contained HTML5 canvas scenes with real physics crash demos Prompts: - A car crashes into a brick wall - A wrecking ball destroys a house - A catapult throws a rock at a castle wall Outputs: Sonnet 5: 15,047 tokens, $0.15 Opus 4.8: 23,063 tokens, $0.58 Sonnet 4.6: 25,824 tokens, $0.39 GPT 5.5: 31,152 tokens, $0.94 Sonnet 5 did as well as Opus 4.8 and GPT 5.5 on all three tests. In the wrecking ball test, it beat Opus 4.8. The cable moves smoothly and every hit connects. In the catapult test, it beat GPT 5.5. The rock always lands inside the wall. Sonnet 5 still needs better detail and graphics. But it used fewer tokens than every other model

  4. Rohan Paul55

    Claude Sonnet 5 在 Intelligence Index 上每任务成本为 $2.29,比 Sonnet 4.6 高约 2 倍,比 Opus 4.8 高约 15%。尽管每 token 单价低于 Opus,但 Sonnet 5 为完成相同任务使用了更多 token,导致总费用更高。标准定价为 $3/百万输入 token、$15/百万输出 token;Anthropic 提供促销价 $2/$10,持续至 2026 年 8 月 31 日,之后恢复原价。目前 Sonnet 5 成本仅次于 Claude Fable 5。

    引用Artificial Analysis@ArtificialAnlys

    Claude Sonnet 5 costs $2.29 per task on the Intelligence Index, a ~2x increase compared to Sonnet 4.6 and ~15% more than Claude Opus 4.8. This is driven entirely by increased token usage, making Claude Sonnet 5 one of the most costly models to run, behind only Claude Fable 5. Our results use standard $3/$15 pricing, however Anthropic is offering a one-third reduction to $2/$10 until September 1

6月30日周二
6月29日周一
  1. MiniMax (official)39

    MiniMax官方转发了Gradient、Parallax和GenericAgent团队的演示结果。他们在本地运行了MiniMax M3(428B参数模型),通过Parallax工具部署在3台Mac上,再由GenericAgent驱动一个约3000行代码的自主智能体,完成了创建5只股票投资组合并写入磁盘的任务。整个过程完全在本地进行,无云端调用、无API费用,数据未离开机器。MiniMax表示这是本地AI未来发展的一个缩影。

    引用Gradient@Gradient_HQ

    A self-evolving agent + a 428B model + 3 Macs = ? Your own AI lab. We ran @MiniMax_AI M3 locally with @tryParallax, right on our desk. Then @GA_agent_ai took over to create a 5-stock portfolio and write it to disk. No cloud. No API bills. Nothing left the machine. Wild to see a ~3K-line agent drive all this with a 400B+ model on local hardware. Thanks to the GenericAgent and MiniMax teams for making local AI feel real.

6月28日周日
  1. PixVerse6

    小企鹅从疾病中康复,本该回归日常,但梦境、星星、微光、旧箱子……奇怪的事情逐渐增多,不过它依然我行我素🐧📷

    引用楽園@dave392750

    病気から回復したぺんぎんちゃん。 いつもの日常に戻ったはずなのに、 夢、星、小さな光、古い箱…… 少しずつ不思議なことが増えてきました。 でも本人は相変わらずマイペースです🐧📷 #pixversecpp @PixVerse_

6月27日周六
  1. OpenRouter49

    OpenRouter 通过 MCP demo 展示智能体实时拉取 DesignArena 的顶级设计模型,并启动三个子代理——GLM-5.2、Opus 4.7、Kimi 2.6——各自生成自画像网页,并排展示供用户挑选。引用推文点出普遍痛点:不同模型各有擅长,但逐一注册、加载凭证、重复跑提示词过于繁琐,致 99% 用户只跟风他人推荐。OpenRouter MCP 提供更便捷的对比方式。

    引用jacky@jjacky

    diff models are good at diff things, but how many of us actually compare them? you sign up for each provider separately, load creds, and run the same prompt over and over it's tedious, so 99% of the time, people just go with what others say is best there's a better way..

6月26日周五