Telegram RegisterThe public register of Telegram
Telegram profile photo for AI 托我给您带个话

Channel

AI 托我给您带个话

@AIWhispers4U

On this record: Topic · Growth · Engagement · Reactions · Posts · Cite this entry

737subscribers

-2 since we began measuring on 7 August 2026

Risers and fallers across the register · movement among entries of Under 1,000.

Register entry

Telegram ID-1001800257036
TypeChannel
Username@AIWhispers4U
CreatedBetween 1 October 2022 and 30 September 2023 — estimated from Telegram’s id allocation, not measured. How this range is calculated.
First recorded7 August 2026
Last confirmed live18 November 2026
Measurements held12
Confirmed unchanged1 time, most recently 18 November 2026
On Telegramt.me/AIWhispers4U

Topic

Technology — a classification, not a measurement. An on-box language model (Qwen3.6-35B-A3B-FP8, prompt version 1) read this channel’s own recent posts on 20 September 2026 and assigned it the closest of 31 fixed categories, at 100% confidence. This is a model’s judgement about what the channel is likely to be about, not a fact this register measured the way a subscriber count or a view count is measured — it can be revised on a later pass, and it carries no weight anywhere else on this page. How this classification works, and the other topics.

Growth

7317397357 August 2026 — 739 subscribers7 August 2026 — 739 subscribers16 August 2026 — 736 subscribers29 August 2026 — 734 subscribers8 September 2026 — 737 subscribers16 September 2026 — 735 subscribers1 October 2026 — 738 subscribers12 October 2026 — 734 subscribers19 October 2026 — 733 subscribers26 October 2026 — 731 subscribers9 November 2026 — 734 subscribers18 November 2026 — 737 subscribers7377 August 202618 November 2026
12 measurements spanning 103 days, net -2. Dots are measurements; the straight line between them is drawn to join them, not to claim we know the path taken in between — snapshots are recorded only when a count changes, so gaps mean “no change observed”, never “interpolated”. The vertical axis spans 730–740 and does not start at zero.
Measurement log — every subscribers count we have recorded
Measured (UTC)SubscribersChange
18 Nov 2026, 12:09737+3
9 Nov 2026, 00:46734+3
26 Oct 2026, 01:51731-2
19 Oct 2026, 08:40733-1
12 Oct 2026, 19:56734-4
1 Oct 2026, 18:17738+3
16 Sept 2026, 11:38735-2
8 Sept 2026, 00:19737+3
29 Aug 2026, 13:54734-2
16 Aug 2026, 19:35736-3
7 Aug 2026, 10:47739no change
7 Aug 2026, 10:39739first reading

Engagement

20 posts held, back to 31 May 2024 — the reader has not yet reached the start of this channel’s public history, so older posts may sit further back, unread. Read across 1 page of Telegram’s post history, 20 posts per page.

Nothing published in the last 30 days. ERR and ER are rolling 30-day measures, so there is nothing to compute — we hold 20 posts for this entry, the most recent from 2 June 2024. An engagement rate over an empty window would be a number about nothing.

Reaction mix

11 reactions across 2 posts, in 2 distinct kinds. The most used accounts for 72.7% of them.

Every reaction kind recorded on the sample, most used first
ReactionCountShareShare, drawn
❤872.7%
👍327.3%

No sentiment is inferred, and none should be read in. This table is ordered by count and by nothing else. Emoji do not carry stable meaning across languages or communities — 🙏 is thanks in one channel and mourning in another — so we publish which ones were pressed and how often, and pass no judgement on what an audience meant by them.

Precision. Telegram publishes reaction counts per emoji and short-forms each one — 4.34K, 1.2M — so any single kind at or above 1,000 reaches us at three significant figures, and only counts below 1,000 are exact. The shares above are ratios of those figures and carry the same error. This is also why the total here can differ slightly from a reaction total printed elsewhere on the page: both are sums of the same rounded parts, taken over samples with different edges.

Coverage. Reactions were read on 5 of the 20 sampled posts in this sample. Summed by Telegram’s own count on each post — not by adding up the per-emoji breakdown above — those same posts carry 11 reactions in total: the kind of figure the paragraph above means by “a reaction total printed elsewhere on the page”.

Measured over the 20 most recent posts we hold, published 31 May 2024 to 2 June 2024, using the newest reading held for each. Telegram Stars are excluded: they are a payment, not a reaction, and they have their own section.

Recent posts

2 Jun 2024, 10:30 UTC≈4,950 views10 reactionsread 7 August 2026

互联网上的大型语言模型的分布式推理和微调存在挑战,特别是上游请求超时。 用户评论强调了分散化人工智能的重要性,分布式技术在加密货币之外的潜力,以及对 GPU 网络的影响。 讨论还涉及到对有用的人工智能工作的证明需求,用于人工智能推理的市场方案,以及 GPU 托管变得更具竞争力的潜力。 一些用户提到了像 Gridcoin 和 BOINC 这样的现有项目,而其他人则幽默地质疑这项技术是否会导致「天网」情景。 提供了主要 arxiv 页面和项目的 GitHub 存储库的链接,供进一步探索。 🔗 https://browse.arxiv.org/html/2312.08361v1 💬 https://news.ycombinator.com/item?id=38843069

❤7👍3

2 Jun 2024, 08:30 UTC≈4,500 views1 reactionsread 7 August 2026

大型语言模型(LLMs)不再仅仅依赖互联网数据进行训练,而是开始融合自定义数据,如注释、人类反馈、使用数据和合成数据,以提升其能力。 公司正在投资专门的工作人员来创建新的训练数据,从而推动 LLM 在互联网模拟之外的性能提升。 这种向自定义数据的转变预计将进一步增强 LLM 的能力。 🔗 https://allenpike.com/2024/llms-trained-on-internet 💬 https://news.ycombinator.com/item?id=40549021

❤1

2 Jun 2024, 06:30 UTC≈3,340 viewsread 7 August 2026

一年来与 LLMs 建设的经验教训,评论可能提供更多深入见解和观点。 🔗 https://eugeneyan.com/writing/llm-lessons/ 💬 https://news.ycombinator.com/item?id=40534293

2 Jun 2024, 04:30 UTC≈2,910 views0 reactionsread 7 August 2026

AI 误导,男子因拨打假冒 Facebook 客服电话被骗。 来自温尼伯的一名男子在拨打一个被 Meta AI 工具认定为真实的假冒 Facebook 客服电话后被骗。 骗子入侵了他的 Facebook 账户,并说服他 IP 地址被黑客入侵,导致他的 PayPal 账户出现未经授权的购买。 男子报案并成功撤销了相关费用。 使用的 AI 工具 Llama 3 并不总是准确的。 男子不再相信 AI,并警告他人不要依赖它来验证事实。 温尼伯警察局正在调查此案。 🔗 https://www.cbc.ca/news/canada/manitoba/facebook-customer-support-scam-1.7219581 💬 https://news.ycombinator.com/item?id=40536860

2 Jun 2024, 02:30 UTC≈1,740 viewsread 7 August 2026

「不精确」的语言模型更小、更快,准确度几乎不减 研究人员正在开发更小、更快、更节能的语言模型,通过将参数精度降低到 1 或 - 1 来实现。 这些 1 位模型在性能和能源效率方面显示出了有希望的结果。 正在使用不同的方法来创建这些模型,它们有潜力直接在手机和针对其特定需求进行优化的定制硬件上运行。 🔗 https://spectrum.ieee.org/1-bit-llm 💬 https://news.ycombinator.com/item?id=40529355

1 Jun 2024, 23:10 UTC≈1,350 views0 reactionsread 7 August 2026

Reddit 文章讨论了一个 AI 创作的现代服装中的火影忍者形象。 评论中有人建议给卡卡西戴上口罩,有人指出最后一个形象其实是我乃悟净,但也可以接受他在火影忍者世界中的存在。 还有人对作品的质量提出了质疑,但也有人认为这个创意还是挺新鲜的。 最后,有人评论说这幅作品是一幅自画像。 #art #ai #creativity 🔗 https://redd.it/1d24ce4

1 Jun 2024, 14:10 UTC960 viewsread 7 August 2026

最近有人注意到 Chatgpt 变得非常啰嗦。 有人抱怨它在回答问题时会重复之前的信息,或者离题。 有人建议在每次对话中使用 「不要啰嗦」 这两个词。 有人发现它在写代码时会重写整个文件,即使只需要修改一点。 有人认为它在夏季会变得啰嗦,冬季会给出简洁回答。 有人说它会忽略对话内容,只是不停地列出清单和维基百科文章。 有人建议调整温度设置以获得简洁回答。 有人发现当要求更详细信息时,它只会给出几个要点。 有人建议使用特定提示来改善信息输出。 有人认为 Chatgpt 在写作时会过于啰嗦,希望能有速度和质量的平衡。 有人建议在提示中设置字符限制。 有人认为 Chatgpt 用于免费更好。 有人想看到更有血性的文字。 #ai #chatgpt 🔗 https://redd.it/1d248o6

1 Jun 2024, 11:10 UTC657 viewsread 7 August 2026

Reddit 文章讨论了 GPT-4o 语音模式的发布时间是否会在 6 月 / 7 月,以及是否有延迟的原因。 有人认为发布被推迟是因为安全团队发出了警报,也有人认为发布时间表可能会延后到年底。 评论中有人指出 Elon Musk 可能会过于乐观地设定时间表,但并不是一个撒谎者。 还有人讨论了 Scarlett Johansen 是否对 Sky 的声音进行了干预,以及对 OpenAI 的不满。 有人提到了 Pi 的声音比 ChatGPT 的声音更自然,也有人提到了希望能够更改声音为其他人的声音。 总体来说,文章讨论了 GPT-4o 的声音模式发布时间以及相关的争议。 #openai 🔗 https://redd.it/1d1ydvp

1 Jun 2024, 08:10 UTC564 viewsread 7 August 2026

总结:讨论的 reddit 文章主要围绕使用 Pony 模型进行图像生成展开,作者 ZootAllures9111 认为在使用高质量的现实 Lora 训练时,Base Pony 的原始、照片、逼真标签在强度为 1.2 时并不那么糟糕。 评论中有人对 Pony 模型的逼真性表示怀疑,也有人分享了使用其他模型的结果。 作者强调可以使用 Pony 模型生成不只是色情内容的图像,并展示了自己的作品。 其他评论包括对图像的评价、对模型的疑问以及关于训练模型时的错误。 整体讨论围绕着图像生成技术展开,探讨了不同模型的适用性和效果。 🔗 https://redd.it/1d1wdj5

1 Jun 2024, 05:10 UTC431 viewsread 7 August 2026

文章讨论了变化的主题,作者表示有时候变化是出乎意料的,有时候变化是如此微妙以至于我们可能都没有注意到。 变化也可以是绝对巨大的,改变生活的。 希望大家都能以希望的方式迎接变化。 评论中有人称赞文章美丽,有人喜欢文章中的图案,有人说喜欢蛾子的图案。 #art 🔗 https://redd.it/1d26wzh

1 Jun 2024, 02:10 UTC365 viewsread 7 August 2026

作者介绍了 Towers 的工作流程,包括使用不同的工具和技术来生成图像,修复和改进图像,以及最终的处理步骤。 评论中有其他用户提出了关于图像解剖结构的问题,作者进行了回应并解释了一些设计选择。 🔗 https://redd.it/1d1tdf9

31 May 2024, 23:10 UTC338 viewsread 7 August 2026

Reddit 文章讨论了 iOS 应用程序上的新语音模式 UI。 作者 none484839 提到他得到了新 UI 的访问权限,但似乎连接到了旧的语音聊天模式。 其他评论者讨论了可能是一个故障,或者新 UI 正在逐渐推出。 有人提到通过代理黑客攻击获得了一些信息,但仍在调查如何使其工作。 还有人提到了使用蓝牙无线耳机时声音质量问题,以及旧的聊天 UI。 #gpt4 🔗 https://redd.it/1d20ni7

Showing the 12 most recent of 20 posts we hold for @AIWhispers4U. View and reaction counts are the latest single reading for each post, not a live figure, and a recent post is still accumulating both. A view count marked ≈ was rounded by Telegram before we ever saw it — t.me prints views in full below 1,000 and to three significant figures above, so ≈1,200,000 means somewhere between 1,150,000 and 1,249,999. Unmarked counts are exact. Text is reproduced from the public post preview and truncated for length.

Cite this entry

A live page changes as we take new readings, so a citation should name the measurement it is based on, not just the URL. The line below cites the subscriber count as measured 18 November 2026 — this entry's latest reading, not the date you are reading this.

“AI 托我给您带个话” (@AIWhispers4U), 737 subscribers as measured 18 November 2026. Telegram Register, tgregister.com/channel/AIWhispers4U.

Full measurement history, CC BY 4.0. Every reading this register holds for this entry, not just the latest one, as a dated, downloadable record: CSV · JSON. Free to use with attribution to tgregister.com. Each file carries its own generation timestamp, which is the figure to cite for exactly when the data was retrieved.