Bonsai-27B-GGUF
lmstudio-community · Kit
A GGUF conversion of the 27B Bonsai model for LM Studio, with a 3.80 GB Q1_0 language file and a 0.93 GB BF16 vision projector. Tested for text, vision and par…
An n8n workflow that posts a motivational quote carousel to Instagram every day. An OpenAI agent writes the caption and three quotes, Blotato turns them into slides and posts them, and seven more networks are one switch away.
Daily quote carousel maker for Instagram is an n8n workflow in the Marketing & SEO 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.
Daily schedule
It starts by itself at 10:00 every day, and there is no form:
A new three-slide quote carousel on your Instagram feed every day
What you need before you start.
Language model
Runs gpt-4o in two agents, of which only the second one’s answer is used. It picks a quote idea and writes the caption, the three quotes and their image prompts.
NeedsOpenAI API key
Slides and publishing
Makes the three quote cards from a slideshow template and an image model, then posts them to the networks you switch on, Instagram first.
NeedsBlotato API key and each connected account’s ID, both in SETUP
Checked by Mindket on 13 Nov 2026
This n8n workflow puts a motivational quote carousel on Instagram every day with nobody at the keyboard. A schedule starts it at 10:00. An OpenAI agent brainstorms 50 quote ideas on personal motivation, picks one at random and writes a caption, three quotes and an image prompt for each. Blotato turns them into three quote cards. The workflow waits three minutes, fetches the finished images and has Blotato post them to Instagram. Seven more networks sit beside it, switched off.
Coaches, creators and small brands that keep an Instagram feed moving with short motivational cards, and would rather read a week of posts than write each one. Skip it if your brand needs its own wording or photos. The quotes come from a model and the pictures from an image generator, and no step asks you to approve them.
| Industry | Who uses it | How they use it | Example |
|---|---|---|---|
| Coaching and wellness | Coaches and counselors | Let the schedule fill the feed with one short card a day | A life coach has a fresh quote card on Instagram every morning. |
| Fitness | Gym owners and trainers | Change the theme in the agent’s prompt from personal motivation to training habits | A gym members’ feed gets a daily card about showing up for the workout. |
| Small business | Shop and café owners | Fill the days between product photos | A bakery posts a card each morning and its product shots on weekends. |
| Personal brands | Creators with several accounts | Switch on the TikTok or Pinterest node beside Instagram | A creator reuses each carousel on Pinterest by turning one node on. |
| Agencies | Social media managers | Keep one copy per client, each with its own theme and account IDs | A manager runs a copy for each client and changes only the theme and the IDs. |
Any n8n that can reach OpenAI and Blotato will run it. A schedule starts it, so nothing needs to reach into your n8n from outside.
scriptGet it saves a zip with workflow.json and Mindket’s setup notes in your downloads folder. Its author puts setup at about 8 minutes.
| Need | Detail |
|---|---|
| n8n | On n8n Cloud or on a server of your own |
| OpenAI | An API key, used by two chat model nodes |
| Blotato | An account with an API key, and your Instagram account connected there |
| Account ID | The ID Blotato gives your Instagram account, which goes into SETUP |
workflow.json.Brainstorm Idea and Brainstorm Idea 2.SETUP. Put your Blotato key between the quotes of blotato_api_key and your Instagram account’s ID in instagram_id.[Instagram] Publish via Blotato and press D to switch it off, so a first run posts nothing.Get Carousel in the run to see the finished images.The Blotato key is not an n8n credential here. It is plain text in SETUP, and the three request nodes and every posting node read it from there.
Once the workflow is published, there is nothing to open. The schedule runs it daily and each run makes one carousel. That is all there is to do. To post to another network, select its node, press D to switch it on and fill its account ID in SETUP. Facebook also needs facebook_page_id, and Pinterest needs pinterest_board_id.
| To change | Where |
|---|---|
| Time | triggerAtHour in Schedule Trigger, the hour of the day |
| Theme | The words personal motivation in the prompt of the agent that feeds SETUP, the one whose answer is used |
| Number of slides | That prompt and the example in Structured Output Parser 2, both written for three |
| Slide look | template, style, captionPosition and text_to_image_model in SETUP |
| Wait for the render | Wait, three minutes |
With the theme left as it is, the agent that feeds SETUP answers in this shape:
{
"caption": "…",
"scenes": [
{ "text": "…", "textToImagePrompt": "…" },
{ "text": "…", "textToImagePrompt": "…" },
{ "text": "…", "textToImagePrompt": "…" }
]
}SETUP adds the Blotato template base/slides/quotecard, the style realistic and the image model replicate/black-forest-labs/flux-1.1-pro. Create Carousel sends it all to Blotato. Three minutes later Get Carousel returns the finished images, and the Instagram node posts them with the caption.
If you are new. n8n draws a workflow as boxes joined by lines, and each box, a node, does one job. Blotato is a service that makes slideshow images and posts to social networks, and your accounts are connected there, so n8n needs only its key. Run it once by hand with the Instagram node off. Each node then shows what it received and sent, which is where a wrong key or an empty account ID turns up.
For experts. Everything outside OpenAI goes to three kinds of Blotato request, and the key travels in a header named blotato-api-key.
| Node | Endpoint | Sends | Gets back |
|---|---|---|---|
Create Carousel | POST /v2/videos/creations | The template, the scenes, the caption position, the style and the image model | An item with the creation’s id |
Get Carousel | GET /v2/videos/creations/<id> | The id | An item with the slides’ imageUrls |
| Posting nodes | POST /v2/posts | The network, the caption, the image addresses and the account ID | Nothing the workflow reads |
Threads, X and Bluesky send the first image as the post and the others as follow-ups in additionalPosts. The other four networks take all three images in mediaUrls. SETUP is a Set node in raw JSON mode, so it is one JSON text with the agent’s answers dropped in by expressions. Keep its quotes and commas when you edit it.
Each run costs money at OpenAI and Blotato, and the workflow posts with no one to approve.
SETUP instead of the other agent’s. Its scenes carry an image prompt and no quote textWait if Blotato is slowisAiGenerated before you switch it onSETUP as plain text and travels with any export of the workflowAn n8n workflow. Every day at 10:00 an OpenAI agent picks a motivational quote, Blotato turns it into a three-slide
carousel, and the carousel goes up on Instagram. Seven more networks are built in and switched off.
workflow.json: the workflow as n8n exports it, 23 nodes, four of them sticky notes. It holds no credential. TheSETUP.README.md: these notes.workflow.json.Brainstorm Idea and Brainstorm Idea 2.SETUP. Put your Blotato API key between the quotes of blotato_api_key and your Instagram account's ID ininstagram_id. Leave the fields for networks you do not use as they are.[Instagram] Publish via Blotato and press D to switch it off for a first run, so nothing is posted yet.Get Carousel in the run: its output holds the finished slide images.No n8n credential is used for Blotato. The key travels in the blotato-api-key header of Create Carousel, Get Carousel
and every posting node, and each of them reads it from SETUP.
Seven posting nodes are switched off: [Facebook] Publish via Blotato, [Linkedin] Publish via Blotato,[Tiktok] Publish via Blotato, [Threads] Publish via Blotato, [Twitter] Publish via Blotato (the one for X),[Bluesky] Publish via Blotato and [Pinterest] Publish via Blotato. To use one, select it, press D to switch it on,
and fill its account ID in SETUP. Facebook also needs facebook_page_id, and Pinterest needs pinterest_board_id.
Threads, X and Bluesky get the first slide as the post and the other two as follow-up posts in the same thread. The
other networks get all three slides at once.
triggerAtHour in Schedule Trigger, set to 10.AI Agent - Specify Quote and Prompt, the agent whose answerStructured Output Parser 2, both written for three.template, style, captionPosition and text_to_image_model in SETUP. The file setsbase/slides/quotecard, realistic, middle and replicate/black-forest-labs/flux-1.1-pro.Wait, three minutes, between the request and the fetch.AI Agent - AI Writes Quote and Prompt, is wired to the trigger, but its output goes nowhere. ItsBrainstorm Idea and Structured Output Parser 1, or move itsSETUP in place of the other agent's. Its scenes hold only an image prompt, with no quote text.Wait if Blotato is slow.SETUP and in every posting node. A caption holding a doubleSETUP, so it travels with any export of the workflow. Take it out before youisAiGenerated is false). The pictures are madeREADME.md 4 KB workflow.json 23 KB
lmstudio-community · Kit
A GGUF conversion of the 27B Bonsai model for LM Studio, with a 3.80 GB Q1_0 language file and a 0.93 GB BF16 vision projector. Tested for text, vision and par…
LibertAIDAI · Kit
Runs DeepSeek-V4.1-Flash-1017 as a 2.9 bits-per-weight EXL3 quantization, a 196 GiB copy for people serving large models on their own hardware. Perplexity stay…
facebook · Kit
Generates short music clips from a text description or an audio prompt, in the small checkpoint of facebook's MusicGen. It suits developers and music hobbyists…
alchaincyf · Skill
Turns a one-line request in your AI agent into a finished design: a launch animation, a clickable app prototype, an editable slide deck, an infographic or a se…
nerijs · Kit
A pixel art style add-on for Stable Diffusion XL that turns text prompts into pixel-style images, with an optional LCM LoRA for faster runs. For game makers, d…
LiconStudio · Kit
Keeps characters, clothing, objects and backgrounds consistent in generated video by feeding the model up to five reference images. A LoRA for LTX-2.5 in Comfy…