Telegram RegisterThe public register of Telegram
Telegram profile photo for Нейроканал

Channel

Нейроканал

@neuro_channel

On this record: Topic · Channels linked by matching posts · Growth · Engagement · What this channel posts · Reactions · Stars · Advertising · Posts · Posts edited after publishing · Posts recorded as deleted · Citations · Handles named that no longer answer · Telegram's recommendations · Cite this entry

12,128subscribers

+1,050 since we began measuring on 5 August 2026

Risers and fallers across the register · movement among entries of 10,000–31,623.

Register entry

Telegram ID-1001585487065
TypeChannel
Username@neuro_channel
DescriptionИскусственный интеллект, нейросети, машинное обучение Разместить рекламу: @tproger_sales_bot Правила общения: https://tprg.ru/rules Другие каналы: @tproger_channels Сайт: https://tprg.ru/site Зарегистрирован в РКН: https://tprg.ru/vcEg
Created25 August 2021 — measured — cross-checked against a third-party dataset (ext.tg_channel)
First recorded6 August 2026
Last confirmed live16 November 2026
Measurements held68
Confirmed unchanged1 time, most recently 16 November 2026
On Telegramt.me/neuro_channel

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 11 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.

Channels linked by matching posts

This channel is in a group of 3 registered channels linked by matching posts, directly or through other channels in the group. This channel may share no posts with some of the channels listed below, and listed channels may share none with each other. Links are found by comparing text fingerprints across every channel on the register.

Which channel, if either, published first is deliberately not shown in this list. “First” could only mean first in our own crawl, which depends on how deep we happened to reach into each channel’s history, not on when the content was actually first posted. Our one hand-check read 45 of the highest-confidence flagged pairs against the live Telegram pages. In 18 of them the live page showed a “Forwarded from” credit that our database had no record of, so reading that pair as an unattributed copy was wrong; the check did not measure who published first. This list is ordered by subscriber count, the same as every other listing on this site, never by which channel we think came first. Near-verbatim matching has several ordinary explanations besides copying — a channel mirroring itself, an unattributed repost, or two channels independently repeating the same wire story — and this measurement cannot tell those apart. How this is measured.

Growth

11,07512,12811,601.55 August 202616 November 2026
68 measurements spanning 102 days, net +1,050. 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 10,917–12,286 and does not start at zero.
Measurement log — every subscribers count we have recorded, most recent 20 of 68
Measured (UTC)SubscribersChange
16 Nov 2026, 09:3112,128+57
12 Nov 2026, 21:5012,071-2
10 Nov 2026, 16:3712,073+2
9 Nov 2026, 03:5812,071+6
7 Nov 2026, 19:5712,065+9
6 Nov 2026, 14:3912,056+10
5 Nov 2026, 13:3712,046+3
4 Nov 2026, 15:5512,043+3
3 Nov 2026, 15:4212,040+3
2 Nov 2026, 17:4912,037+21
1 Nov 2026, 18:3912,016+13
31 Oct 2026, 15:1512,003+6
30 Oct 2026, 17:3011,997+1
29 Oct 2026, 16:5611,996+16
28 Oct 2026, 16:5711,980+13
27 Oct 2026, 15:4711,967+19
26 Oct 2026, 12:1511,948+6
24 Oct 2026, 14:1611,942-4
23 Oct 2026, 17:3611,946+7
22 Oct 2026, 15:3811,939first reading

Engagement

203 posts held, back to 31 July 2026 — the reader has not yet reached the start of this channel’s public history, so older posts may sit further back, unread. Read across 75 pages of Telegram’s post history, 20 posts per page.

ERR · 30 days
21.6%
avg views ÷ 12,128 subscribers
Avg views / post
2,620
76 posts measured
Reaction rate
0.469%
reactions ÷ views · ER floor
Posts in window
76
of 203 held

ERR is average views per post over the last 30 days divided by subscribers, the definition TGStat uses, so this figure is comparable with the one you will see elsewhere. It falls structurally as a channel grows: a high ERR on a small channel and a low one on a large channel describe reach mathematics, not quality. We publish the figure and the sample it came from and pass no verdict on it.

ER is defined industry-wide as (forwards + reactions + comments) ÷ views — note the denominator is views, not subscribers. Telegram’s public web preview carries views and reactions but not forward or comment counts, so the reaction rate above is the reactions term only and is therefore a floor: the true ER for this channel is higher by an amount we have not measured and will not estimate. It is computed over the 75 of 76 measured posts that carry a reaction reading, and over those same posts' views.

What these figures were computed from
WindowRolling 30 days · latest post in window 16 November 2026
Posts held203 (31 July 2026 – 16 November 2026)
Views total199,210
Reactions total922
Forwards / commentsnot exposed by the public surface — not measured, not estimated
Readings taken17 Nov 2026, 11:28 UTC

Views are a single reading per post, taken at the time above. A post published in the last day or two is still accumulating views, which pulls the 30-day average down slightly. That is a property of the standard definition rather than a fault in it, so we keep the definition rather than “correcting” the number into something nobody can reproduce.

Precision. Telegram publishes view counts on its public widget in short form — 8.12K, 3.7M — so any reading at or above 1,000 reaches us rounded to three significant figures, and only counts below 1,000 are exact. Averages and rates derived from them are shown to the same precision rather than to the unit: a figure like 3,701,250 would assert digits nobody measured.

Reaction counts are published per emoji and rounded the same way, so a total below 1,000 is exact and a larger one is a sum that may carry a rounded component from each emoji above 1,000. Because it is a sum, it does not look rounded — read a large reaction total as three significant figures per contributing emoji rather than as the figure it prints.

What this channel posts

Photos
≈987
Videos
≈261
Links
≈2,190

Lifetime counters from Telegram’s own channel header, read 17 November 2026 — not the date at the top of this page, which is when the subscriber count was last read. A count marked ≈ was rounded by Telegram before we ever saw it — t.me prints these counters in full below 1,000 and to three significant figures above, so ≈142,000 means somewhere between 141,500 and 142,499.

Video runtime
7m 18s
Average length
34s

Measured directly from 13 videos with a duration reading, out of the posts we hold for this channel — not this channel’s whole posting history, only the sample this register has actually read. An exact reading to the second, taken from the post itself rather than from Telegram’s own rounded chrome, so it carries no ≈ mark.

Reaction mix

2,243 reactions across 196 posts, in 35 distinct kinds. The most used accounts for 26.5% of them.

Every reaction kind recorded on the sample, most used first
ReactionCountShareShare, drawn
❤59426.5%
👍39017.4%
🔥37816.9%
✍1466.51%
😁1386.15%
🤣893.97%
⚡683.03%
🌚522.32%
🤔522.32%
😱401.78%
🍌361.60%
💯341.52%
🎉281.25%
👎281.25%
👀241.07%
😭220.981%
🥱190.847%
🤯170.758%
😢150.669%
🆒110.49%
15 further kinds622.76%

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 200 of the 203 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 2,279 reactions in total: the kind of figure the paragraph above means by “a reaction total printed elsewhere on the page”.

Measured over the 203 most recent posts we hold, published 31 July 2026 to 16 November 2026, using the newest reading held for each. Telegram Stars are excluded: they are a payment, not a reaction, and they have their own section.

Telegram Stars

Stars received
4
across the posts below
Posts paid on
3
of 203 we hold a reading for · 1%
Most on one post
2
single highest reading

A paid reaction is a reader spending Telegram Stars — bought with money — on a post by @neuro_channel. Telegram publishes the count on the public post preview alongside ordinary reactions, and this register reads it there. It is the only figure on this site that measures money moving rather than attention.

Stars are not reactions, and the two are never added. They are rendered in the same strip on Telegram and counted in the same shape, but one is a tap and the other is a purchase. The reaction totals and the engagement rate elsewhere on this page exclude every figure in this section, and no rate here is computed against a reaction count.

This is not revenue, and we publish no currency figure. What a Star costs a reader and what it pays a channel are different numbers, Telegram takes a share we cannot observe, and the terms have changed. Converting a Star count into money would be an estimate dressed as a measurement, so the count is where we stop.

Counted over the 203 most recent posts we hold for this entry, published 31 July 2026 to 16 November 2026. Star counts above 1,000 reach us in Telegram’s short form and carry the same three-significant-figure rounding as everything else on this page.

Advertising

Ad load
0.985%
2 of 203 posts carry an ad marker
Regulatory tokens
2
posts carrying an erid · 2 distinct tokens
Median views · ads
1,440
over 2 measured posts
Median views · rest
1,830
over 201 measured posts

An ad marker, not a judgement about a post. A post is counted here because it carries one of two explicit markings: an erid token, which Russian law has required on paid placements since 2022 and which is issued against a specific advertising contract, or a #реклама / #ad hashtag in the body, which is the channel declaring it itself. The first is documentary; the second is a self-declaration and is weaker. No classifier reads the text and decides — nothing on this site guesses that a post is an advertisement.

This is a floor, and it can only ever be a floor. A channel that runs paid placements without marking them produces no marker for us to count, and an unmarked ad is indistinguishable from an ordinary post on the public surface. The ad load above therefore means “the share of posts that declared themselves”, never “the share of posts that were paid for”. A low figure is not evidence of a channel that runs few ads.

Both figures are medians, and no ratio between them is published. Each is a view reading that actually occurred on a post, picked by percentile_disc rather than averaged, so one viral post cannot move it and no interpolated value is invented between two readings. The sample on one side is under five posts, which is too thin to compare. The two figures are shown side by side with the count behind each, and deliberately not divided into a headline like “ads get x% fewer views” — an arithmetic that is easy to print and, at this sample size, means nothing.

Advertising tokens recorded on this entry
eridPostsFirst seenLast seen
2VtzqwBPKej120 August 202620 August 2026
2W5zFHuRC7U111 November 202611 November 2026

A token repeated across several posts is one advertising contract placed more than once, which is what the identifier is for. The strings are reproduced exactly as they appeared in the post or in its click-through URL and are not validated against any registry — we record the marker a channel published, and whether it resolves to a real contract is a question for the register that issued it.

Measured over the 203 most recent posts we hold, published 31 July 2026 to 16 November 2026. Views are the latest single reading held for each post, and any reading at or above 1,000 is rounded by Telegram to three significant figures.

Recent posts

Telegram’s own ‘edited’ marker: 18 of 76 posts held from the last 30 days (23.7%) carried it when this register last read them. The marker says a post was changed after publishing, not what it said before — this register holds the earlier wording only where an edit fell between two of its own reads (see the edit ledger and how it is measured).

16 Nov 2026, 16:44 UTC≈1,300 views6 reactionsread 17 November 2026
Photo

В открытом проекте smolmaxxing модель на 80 млрд параметров запустили на телефоне. На OnePlus Nord CE 2 2022 года Qwen3-Next-80B-A3B выдаёт 2,6–3,1 токена в секунду при пике около 5 ГБ памяти, Qwen3.6-35B-A3B 4,9–5,7. Нижние цифры получены с точной маршрутизацией, верхние с небольшим смещением роутера в пользу экспертов, которые уже в памяти: для 80B-модели это стоит 0,5% перплексии. Замеры сделаны с контекстом 1024 …

👍5🙈1

16 Nov 2026, 16:21 UTC≈1,390 views4 reactionsread 17 November 2026
Photo

DeepSeek открыла в API веб-поиск с цитатами, поиск по ходу рассуждений, разбор PDF и офисных документов, чтение ссылок, распознавание и синтез речи. В запросе к deepseek-flash поиск и чтение ссылок включаются инструментами deepseek_web_search и deepseek_read_url, а ищет и загружает страницы сервер. Работает в Chat Completions, Responses и Messages. Инструменты оплачиваются сверх токенов: поиск и загрузка страниц сто…

❤4

16 Nov 2026, 15:58 UTC≈1,370 views2 reactionsread 17 November 2026

Вышел llama․cpp 1.0.0, а вместе с ним ggml 1.0.0. Версия 1.0.0 означает обещание: ломающие изменения публичного C API и ABI теперь меняют мажорную версию, а минорная растёт от новых моделей, бэкендов и функций. Перед этим API почистили: из llama.h убрали 35 устаревших функций и флаг --defrag-thold. Из нового: KV-кэш в FP8, кэш экспертов MoE в оперативной памяти, поддержка Kimi-K3.1-Pro и эндпоинт /v1/audio/speech. В…

❤2

16 Nov 2026, 15:36 UTC≈1,450 views5 reactionsread 17 November 2026
Photo

Тибо пишет, что команда Codex, похоже, починила ошибки «model at capacity», которые многих достали. Правда, говорит об этом осторожно: «возможно, действительно починили». У меня проблем не наблюдается в последнее время. @neuro_channel

❤5

16 Nov 2026, 11:31 UTC≈1,570 views10 reactionsread 17 November 2026

Топ трендов HuggingFace за неделю — что приехало в опенсорс TL;DR Новое: Edge0-145B запускает большую MoE на одной игровой видеокарте. LongCat-2.5 на 1,6 трлн параметров под MIT. Inkling 2 Small получила веса. ZONOS3-Mini озвучивает текст прямо в браузере. Haru-4B слушает и говорит одновременно. MiniMax H3 Image рисует картинки на 16 Мп. Mirage разбирает картинку на телефоне за 74 мс. Держится: Qwen3.8-27B, LTX-2.…

👍8❤2

16 Nov 2026, 11:06 UTC≈1,600 views8 reactionsread 17 November 2026
Photo

Alibaba вывела Qwen4-Max из превью: модель появилась в API с оплатой по токенам, $1,50 за миллион входных токенов и $4,50 за выходные, на OpenRouter цена та же. В превью, про которое я писал, её давали только по подписке Token Plan. Artificial Analysis уже добавила модель в свой индекс: 43 балла. Это выше DeepSeek V4.1 Flash (33) и GLM-5.3 (32), ниже Kimi K3.1 Pro (46) и далеко от GPT-6.1 Sol (55), Claude Opus 5.5 (…

👍7🔥1

15 Nov 2026, 06:32 UTC≈1,880 views4 reactionsread 17 November 2026
Video

Что нового у Claude: Projects в Claude Code стали проактивнее, десктоп получил темы, версия 2.1.325 перевела Bash в песочнице на обычные разрешения, а сервис Managed Agents можно развернуть в своей инфраструктуре. 🔘 Projects в Claude Code обновили: координатор, который ведёт параллельные треды, стал проактивнее, сам отвечает на вопросы в чате проекта и сообщает статус задач. 🔘 Claude Code 2.1.325 меняет умолчания: …

👍3😭1

15 Nov 2026, 04:06 UTC≈1,830 views4 reactionsread 17 November 2026
Photo

Artificial Analysis обновила индекс кодинг-агентов до v2.1: Terminal-Bench подняли до 4.1, а FrontierCode до 2.0, в нём оценивают, принял бы правку мейнтейнер репозитория. По данным AA, из-за нового FrontierCode оценки упали в среднем на 5 баллов. Наверху с 58 баллами Claude Fable 5.2 на xhigh и Claude Opus 5.5 на max, обе в Claude Code. У Fable 5.2 на high те же 58 баллов, а задача стоит $7,5 против $15,5 на xhigh …

👍3❤1

14 Nov 2026, 07:03 UTC≈1,950 views11 reactionsread 17 November 2026

На OpenRouter приехала новая пачка моделей. Claude Fable 5.2, Gemini 4 Carbon и Hydrogen, Gemini Decision 1 и Inkling 2 Small я разбирал отдельно; у Inkling с тех пор появились веса под Apache 2.0, как и обещали. Здесь остальное. 🔘 Muse Ember, самая маленькая модель лаборатории суперинтеллекта Александра Вана, сделана для рутины агентов: вызовы инструментов, подзадачи кодинг-агентов, роутинг, извлечение данных, разб…

👍9❤1🍌1

13 Nov 2026, 12:35 UTC≈2,050 views11 reactionsread 17 November 2026
Photo

OpenAI и Yubico продают пользователям ChatGPT набор из двух ключей YubiKey за $68 вместо $126. В нём YubiKey C NFC, который носят на связке и прикладывают к телефону, и YubiKey C Nano, который предлагают постоянно оставлять в USB-C ноутбука. В одни руки до 3 наборов, купить можно в США, Великобритании и Европе, по справке OpenAI. Про скидку напомнил сотрудник OpenAI. Ключи подходят для режима Advanced Account Securi…

❤8🤔3

13 Nov 2026, 11:33 UTC≈2,010 views4 reactionsread 17 November 2026

Что нового у харнессов: в Claude Code появился Code Mode, в Cursor с агентом можно говорить голосом, а Qwen Code 1.0 вышел в превью и ставится без отдельной установки Node.js. 🔘 Claude Code 2.1.323 добавляет экспериментальный Code Mode, который включается отдельно: Claude пишет JavaScript, который сам вызывает инструменты. Сводки фоновых субагентов, про которые я писал, выключаются целиком через CLAUDE_CODE_DISABLE_…

❤4

13 Nov 2026, 10:10 UTC≈1,920 views6 reactionsread 17 November 2026

Alibaba открыла превью Qwen4-Max, первой модели нового поколения Qwen4. В API её можно взять только по подписке Token Plan, оплаты по токенам нет. Превью идёт до 19:00 мск 25 ноября, потом эту версию отключат. Модель мультимодальная, по умолчанию рассуждает, контекст 1 млн токенов. API совместим с OpenAI и Anthropic, так что её можно подключить в Claude Code, Codex или Qwen Code. Подписка стоит от $6 в месяц, новым …

❤6

Showing the 12 most recent of 203 posts we hold for @neuro_channel. 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.

Posts edited after publishing

@neuro_channel edited 1 post after it first published — the same permalink now carries different wording than the one this register originally read, caught because our own crawl held a copy of the earlier text.

An edit is not deception. Typo fixes, price updates and corrections look exactly like this too — this register can tell you the wording changed and when, not why. How this is measured.

First edit seen
19 August 2026
Most recent edit
19 August 2026

Posts recorded as deleted after publishing

We recorded 4 posts on @neuro_channel as deleted — posts this register held that were missing from the channel’s public page on two separate reads, taken from different network addresses at least an hour apart.

A deletion is not wrongdoing. This register can see that a post it held is no longer on the channel’s public page, and when — not why. Expired ad placements, reposts, rotating links and corrections all look the same. This is a count of recorded detections, not an exact total: it misses deletions we could not see, because only a post we held and later re-read inside the page can be seen to go, and it may include a small number of wrong records. How this is measured.

First seen missing
1 September 2026
Most recent seen missing
16 November 2026
Median time before seen missing
4 days
Seen missing in the last 30 days
3

Forward network

Built only from forwarded posts we have actually read, on both sides. Coverage is early and deliberately incomplete: a missing link means we have not read the post that would prove it, never that the relationship does not exist. Counts are distinct forwarded posts observed, so they only ever go up as we read more.

Mentions

Named by 9 registered channels — every channel on the register whose own posts have named this one, by its current username or any other username it currently holds, merged from two separately captured readings of the same fact so a namer caught by only one of them is not missed and a namer both caught is not counted twice. A username this channel has since dropped is not matched — that handle may belong to someone else now, and crediting today’s namer to yesterday’s owner would misattribute it.

Names

Channels on the register whose handles appear in this channel's posts.

A mention is a weaker signal than a forward and is counted separately for that reason — naming a channel is not republishing it, and a handle in a post body is easy to place deliberately. The post counts beside each row below are distinct posts in which the handle appeared, from posts we have read on both sides — the “Named by N registered channels” figure above is a different count, of distinct NAMING CHANNELS rather than posts, and is not the sum of the rows under it.

Appears in Telegram’s recommendations for other channels

The reverse of the list above, and a different kind of signal. This does not require this channel to have ever been asked about directly — each row below is a channel we DID ask Telegram about, whose Telegram-generated list happened to include this one. A channel can appear here with an empty list above it, because being named by someone else’s query is independent of having been queried itself.

Data Science | Machinelearning [ru]
@devsp · 19,622
Telegram ranks this channel #18 of 91 here — alongside 90 others — read 30 September 2026
Типичный программист
@tproger · 77,999
Telegram ranks this channel #26 of 95 here — alongside 94 others — read 19 August 2026
NOP::Nuances of Programming
@nuancesprog · 55,879
Telegram ranks this channel #37 of 84 here — alongside 83 others — read 23 August 2026
Sber AI
@SberAIScience · 26,376
Telegram ranks this channel #49 of 95 here — alongside 94 others — read 10 September 2026
Веб-страница
@tproger_web · 22,114
Telegram ranks this channel #54 of 96 here — alongside 95 others — read 20 September 2026
Zen of Python
@zen_of_python · 18,614
Telegram ranks this channel #57 of 91 here — alongside 90 others — read 5 October 2026
Machinelearning
@ai_machinelearning_big_data · 274,992
Telegram ranks this channel #60 of 95 here — alongside 94 others — read 10 August 2026
[404] — программирование
@procode404 · 40,018
Telegram ranks this channel #73 of 83 here — alongside 82 others — read 30 August 2026
Pythonist.ru - образование по питону
@pythonist_ru · 23,857
Telegram ranks this channel #75 of 82 here — alongside 81 others — read 17 September 2026

This channel appears in 9 seed channels' Telegram-generated recommendation lists in total. Each is Telegram’s list for THAT channel, not this one — see how this is measured.

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 16 November 2026 — this entry's latest reading, not the date you are reading this.

“Нейроканал” (@neuro_channel), 12,128 subscribers as measured 16 November 2026. Telegram Register, tgregister.com/channel/neuro_channel.

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.