Summarize
Summarize
Section titled “Summarize”Condense long text into a short summary, running locally in your browser with a summarization model (e.g. DistilBART CNN on transformers.js). Private, offline, no cost.
What it does
Section titled “What it does”Runs the text through an on-device summarization model with a length preset. Long input is handled for you: it is split into chunks on sentence boundaries (about 2,500 characters each), each chunk is summarized, and — if the joined summaries are still long — they are summarized once more.
When to use it
Section titled “When to use it”Get the gist of an article, a long email thread, meeting notes, or a scraped page before storing it, notifying someone, or feeding it to another step.
Inputs and settings
Section titled “Inputs and settings”| Setting | Notes |
|---|---|
| Text | The text to summarize. Any length — long input is chunked automatically. |
| Length | Short (≈ 20–60 words), Medium (≈ 60–130, default), or Detailed (≈ 130–250). |
| Model | An installed summarization model. Only summarization models are listed here. |
| Min tokens (advanced) | Overrides the preset’s minimum length. |
| Max tokens (advanced) | Overrides the preset’s maximum length. |
Lengths are counted in tokens — roughly one per word in English.
Outputs
Section titled “Outputs”Returns { summary, words }:
summary— the summary text.words— its word count.
Troubleshooting
Section titled “Troubleshooting”- “Choose a summarization model.” — install one from the picker or the Local AI page.
- “Provide some text to summarize.” — the Text field resolved to empty.
- Slow on very long text — every chunk is a separate model run, done one after another. Summarize only the part you need, or use a shorter preset.
- Summary too short or too long — change the Length preset, or set exact Min/Max tokens.
Related nodes
Section titled “Related nodes”- Translate, Classify Text
- Generate Text — free-form text-to-text transforms.
- Summarization Chain — summarize with a chat model instead.