DeepSeek-R1-GGUF
unsloth · Kit
GGUF copies of the 671B DeepSeek-R1 reasoning model, with 1.58-bit and 2-bit dynamic quants, for running it in llama.cpp on a local machine with enough memory.
An n8n workflow that plays a song when you ask for it in Telegram. An OpenAI model picks out the title and the artist, Spotify finds the track and plays it next, and the bot replies with what is playing.
Telegram song requests played on your Spotify is an n8n workflow in the Productivity category, listed by Mindket. It is free. This is version 1, last updated on . Download it on this page or copy its JSON, then import it into an n8n canvas. Nothing runs until you activate it in n8n.
A Telegram message
It starts each time you or another listed user sends a message to your bot, and there is no form:
The song plays on your Spotify device, and a Telegram reply names it
What you need before you start.
Chat app
Delivers your request to the workflow and sends the reply.
NeedsA bot token from BotFather, as a Telegram credential
AI model
Turns your words into a title and an artist.
NeedsAn OpenAI API key and some credit
Music player
Finds the track, queues it and plays it.
NeedsA Premium account and a Spotify OAuth2 credential from your own Spotify app
Checked by Mindket on 13 Nov 2026
This n8n workflow lets you control your music from a Telegram chat. You write the name of a song, or describe it if you forget the name. An OpenAI model turns the text into a search line with the title and the artist, and Spotify finds the track, puts it next in the queue and plays it. The bot then replies with the title, the artist and the album that now play. Thirteen nodes do the work, beside two sticky notes.
Spotify Premium users who like to send a song request from their phone, and small venues or offices with a shared speaker and one trusted person who picks the music. Skip it if you do not have Premium, because Spotify only lets Premium accounts control playback. Premium is required.
| Industry | Who uses it | How they use it | Example |
|---|---|---|---|
| Hospitality | Cafe owners | Change the music from behind the counter | An owner types “something calm by Norah Jones” and the next song starts on the shop speaker. |
| Office | Team leads | Play a song for the team from a chat | A lead sends the title of a favorite song before a Friday meeting. |
| Home | Families | Ask for a song without opening the app | A parent writes a song title while cooking, and it plays on the kitchen speaker. |
| Events | Small event hosts | Take requests from one trusted helper | A host lets the DJ’s assistant send requests from a phone during a party. |
You need some n8n: import a file, add three credentials, make a Spotify app and fill in one number.
Any n8n with a public HTTPS address can run it, because Telegram sends each message to that address.
Download saves a zip with workflow.json and Mindket’s setup notes in your downloads folder. You need a Telegram bot token, an OpenAI API key and a Spotify Premium account with a Spotify app.
| Need | Detail |
|---|---|
| n8n | On n8n Cloud or on a server with a public HTTPS address |
| Telegram | A bot and its token, which you get from BotFather inside Telegram |
| OpenAI | An API key and some credit |
| Spotify | A Premium account, an app with its client ID and secret, and Spotify open on a device |
workflow.json.Search track, add a Spotify OAuth2 credential with that client ID and secret, and connect your account. Select the same credential in Add song, Next Song, Resume play and Currently Playing.Ask OpenAI about a track, add an OpenAI credential.Telegram Trigger, add a Telegram credential with your bot token. Select it in Return message to Telegram too.Telegram Trigger, press Test workflow and send a message to your bot. Copy the id inside from in the output, put it in the field and save.play Yesterday by the Beatles to your bot.Once the workflow is on, write a song to the bot at any time. Each message starts one run, and one reply comes back. Spotify must be open on a device, and the bot ignores everyone but the user numbers in the field.
| To change | Where |
|---|---|
| What the AI looks for | The prompt in Ask OpenAI about a track |
| The model | The model name in Ask OpenAI about a track |
| The “not found” text | Not found error message |
| The “now playing” text | Define Now Playing |
| Who may use the bot | Restrict to User IDs in Telegram Trigger |
A cafe owner plays music from a Telegram chat.
During the lunch rush the owner writes “the song with the blue moon, a slow one”. The AI turns it into a search line, Spotify finds a track and the queue jumps to it. The bot answers with the title, the artist and the album. The guess can be wrong for a vague description, so the owner checks the reply and sends a clearer request if the song is not the one they meant.
If you are new. n8n draws a workflow as boxes joined by lines, and each box, a node, does one job. Telegram is a chat app, and a bot is an account that a program runs. This workflow reads your message, asks an AI which song you mean and tells Spotify to play it. Try it with your own user number in the field before you turn it on.
For experts. Every branch ends in the same text step, so a Spotify error reaches the chat as plain text, and one join gives the chat number.
| Node | What it does | What to know |
|---|---|---|
Ask OpenAI about a track | Turns the text into track:… artist:… | It names gpt-6-luna, and the prompt asks for that one line only |
Search track | Asks Spotify for one match | Always Output Data is on, and its error output goes to the text step |
Add song | Puts spotify:track:<id> in the queue | It needs an active device and Premium |
Next Song | Skips to the queued track | It cuts off the song that plays now |
Resume play | Presses play | It ignores its own error, because Spotify refuses the call when it already plays |
Define Now Playing | Writes the title, the artist and the album | It uses the field error, so the text step is shared |
Merge | Joins the text with the Telegram message | It supplies the chat number to the last node |
The original copy named a model that OpenAI is retiring, and it let anyone who found the bot control the owner’s music.
gpt-4o-mini. OpenAI’s deprecations page, read on 12 November 2026, lists it for shutdown on 14 January 2027, with gpt-6-luna as its replacement. This copy names gpt-6-lunaTelegram Trigger is now on node version 1.2, where Restrict to User IDs is enforced. It holds the placeholder YOUR_TELEGRAM_USER_ID, so the bot ignores everyone until you put your own number therelocalhost as a redirect address_ or * can make Telegram refuse the replyAn n8n workflow that plays a song when you ask for it in Telegram. You write the name of a song, or describe it if you forget
the name. An AI model picks out the title and the artist, Spotify finds the track, puts it next in the queue and plays it, and
the bot tells you what is playing.
workflow.json: the workflow as n8n exports it, 13 nodes and two sticky notes. It holds no credential.README.md: these notes.Ask OpenAI about a track asked for gpt-4o-mini. OpenAI's deprecations page, read on 12 November 2026, lists that model forgpt-6-luna as its replacement. This copy names gpt-6-luna. The node has no tools, soTelegram Trigger is now on node version 1.2YOUR_TELEGRAM_USER_ID. Until you put your own number there, the bot ignores everyone.Return message to Telegram has the option "Append n8n Attribution" off. On node version 1.1 and later, n8n adds the line "Thisworkflow.json.Search track, add a Spotify OAuth2 credential with that client ID and secret, and connect your Spotify account. SelectAdd song, Next Song, Resume play and Currently Playing.Ask OpenAI about a track, add an OpenAI credential with your API key.Telegram Trigger, add a Telegram credential with your bot token. Select the same credential in Return message to Telegram.Telegram Trigger, press Test workflow and send a message toTelegram Trigger, copy the id inside from. Put it in the field and save.play Yesterday by the Beatles to your bot.Telegram Trigger starts the run when you write to your bot.Ask OpenAI about a track turns your text into a search line, such as track:Yesterday artist:The Beatles.Search track asks Spotify for one match. If checks that the match has an id. Without one, Not found error message writesAdd song puts the track in the queue, Next Song skips to it and Resume play presses play. Resume play ignores its ownCurrently Playing reads what plays, and Define Now Playing writes one line that holds "Now playing", the title, the artist and the album.Message parser keeps the text of whichever branch ended, and Merge joins it with the Telegram message to get the chat.Return message to Telegram sends the text.Ask OpenAI about a track.Not found error message.Define Now Playing.Ask OpenAI about a track.Telegram Trigger. You can list several numbers with commas.Next Song jumps to the queued track, so the current one is cut off. Spotify'slocalhost as a redirect address. On an n8n that runs only on your own computer, use the address127.0.0.1 or put n8n behind an HTTPS address._ or * can make TelegramNOTICE.md 581 B README.md 6 KB workflow.json 10 KB
unsloth · Kit
GGUF copies of the 671B DeepSeek-R1 reasoning model, with 1.58-bit and 2-bit dynamic quants, for running it in llama.cpp on a local machine with enough memory.
prithivMLmods · Kit
Provides GGUF files of Axion-1.0-35B, a 35B mixture-of-experts model with about 3B active parameters aimed at scientific reasoning, information seeking and too…
Phips · Kit
Upscales low-resolution images with a super-resolution network that trains on one consumer GPU and exports to ONNX. For people who train or run image upscalers.
ACE-Step · Kit
Generates music from text with a 4B diffusion decoder, 50 steps and guidance control, aiming for higher audio quality than the base. For creators with a GPU of…
op7418 · Skill
Skill that makes single-file HTML slide decks, slide illustrations and platform covers in two visual styles, with a presenter mode for rehearsal. For speakers…
zarazhangrui · Skill
Builds HTML presentations from scratch or from a PowerPoint file and shows visual style previews to pick from, for people who want web slides with no CSS or Ja…