Key Takeaways
- Text Generator is free and open-source; the repository's own manifest.json confirms the MIT License
- Built by developer Noureddine Haouari, with documentation at text-gen.com and docs.text-gen.com
- Built-in named providers: OpenAI, Anthropic, Google Gemini (Gemini-Pro), and HuggingFace
- Local models are reached through the generic "Default (Custom)" provider slot, pointed at any OpenAI-compatible base URL — this is how it connects to Ollama, LM Studio, or similar local servers
- GitHub repository (github.com/nhaouari/obsidian-textgenerator-plugin) shows 1,985 stars as of September 12, 2026, per the GitHub API
- Actively maintained: a commit as recently as August 2026, and a tagged release, v0.8.7, published April 27, 2026
- Template engine driven by note frontmatter, plus a community hub for discovering and sharing templates
📍 In One Sentence
Text Generator is a free, MIT-licensed Obsidian plugin that inserts AI-generated text into notes from reusable templates, with local-model access via a generic OpenAI-compatible custom-provider slot rather than a named Ollama integration.
💬 In Plain Terms
Instead of a chat window you type into, Text Generator runs a template — a saved prompt with frontmatter settings — against your selected text or note and writes the result straight into the document. If you point its custom-provider setting at a local server address instead of a cloud API, it runs entirely against a model on your own computer.
What Is Text Generator?
Text Generator is a community plugin for Obsidian, the notes app, described on its own GitHub repository as "a versatile plugin for Obsidian that allows you to generate text content using various AI providers." It is built by developer Noureddine Haouari and is not an official Obsidian product. Its design centers on templates rather than an ongoing conversation: you configure a prompt once, and reuse it against different notes or selections.
- Core function: generate ideas, titles, summaries, outlines, and full paragraphs using reusable templates, inserting the result directly into the active note
- Local inference: no bundled engine of its own — for local use, its generic custom-provider setting connects over HTTP to a separately running local server (such as Ollama or LM Studio) that exposes an OpenAI-compatible API
- Developer: Noureddine Haouari, with a companion site at text-gen.com and documentation at docs.text-gen.com
- Canonical repository: github.com/nhaouari/obsidian-textgenerator-plugin, MIT-licensed
- Not to be confused with text-generation-webui (also known as "oobabooga"), an unrelated standalone web UI for running local models — the shared "text generation" wording is coincidental
Background and Development Activity
Text Generator is an independent, open-source community plugin distributed through Obsidian's Community Plugins catalog. Unlike some comparable plugins, its GitHub repository shows ongoing activity: a commit as recently as August 2026, and its latest tagged release, v0.8.7, published April 27, 2026, per the repository's own manifest.json and releases page.
- 1Community plugin release
Why it matters: Developer Noureddine Haouari published Text Generator through Obsidian's Community Plugins catalog as an open-source, template-driven text-generation tool. - 2Multi-provider support added
Why it matters: The plugin added named support for OpenAI, Anthropic, Google Gemini, and HuggingFace, alongside a generic custom-provider option for any other OpenAI-compatible endpoint, including local servers. - 3Template engine and community hub added
Why it matters: Frontmatter-based template configuration and a hub for discovering and sharing community-built templates turned the plugin from a single-prompt tool into a reusable template library. - 4v0.8.7 released — April 27, 2026
Why it matters: The most recent tagged release shown on the repository's [Releases page](https://github.com/nhaouari/obsidian-textgenerator-plugin/releases) as of this review, with recent release notes documenting added support for newer OpenAI model families and thinking-model configurations.
What Can You Do With Text Generator?
Text Generator's feature set centers on reusable, template-driven generation rather than an open-ended chat interface, per its own README and documentation.
- Template engine — save a prompt as a reusable template, configured through frontmatter (YAML metadata at the top of a note), so the same generation task can be run repeatedly without retyping the prompt
- Direct insertion into notes — generated text (ideas, titles, summaries, outlines, or full paragraphs) is written straight into the active note or a selection, rather than displayed in a separate chat panel you copy from
- "Considered Context" — flexible prompt-context options that let a template pull in surrounding note content, selected text, or other configured context sources
- Community template hub — discover and share templates other users have built for specific generative AI use cases, rather than writing every prompt from scratch
- Named cloud providers — OpenAI, Anthropic, Google Gemini (Gemini-Pro), and HuggingFace, each configured with your own API key
- Generic custom provider ("Default (Custom)") — a configurable base URL and model name that accepts any OpenAI-compatible endpoint, which is the documented path to a fully local server such as Ollama or LM Studio
- Frontmatter-driven configuration — per-template settings (model, temperature, provider) are set in a note's frontmatter, letting different templates use different providers or models
Usage Examples: Three Ways to Use Text Generator
These are concrete workflows built from the plugin's own documentation, not hypothetical use cases.
Point the custom provider at a local Ollama server
With Ollama running locally and a model pulled, open Text Generator's settings, select the "Default (Custom)" provider, and enter Ollama's OpenAI-compatible base URL (http://localhost:11434/v1) along with the model name you pulled. Running a template afterward sends the request to your local Ollama instance instead of a cloud API.
Generate an outline from a template
Select a community or custom template built for outlining, run it against a note containing rough notes or a topic heading, and Text Generator inserts a structured outline directly into the note — using whichever provider (cloud or local) that template's frontmatter specifies.
Summarize a long note with "Considered Context"
Open a lengthy note, run a summarization template, and Text Generator's "Considered Context" option pulls in the relevant surrounding content automatically, then writes a summary into the note without you needing to manually copy the source text into a prompt.
Get Text Generator
Text Generator installs as a plugin inside Obsidian, not as a standalone application — Obsidian (available for macOS, Windows, Linux, iOS, and Android) must already be installed. There is no separate installer per operating system; the same plugin package works across every platform Obsidian supports.
Community Plugins browser (recommended)
- How to Install:
- In Obsidian, open Settings → Community plugins → Browse, search for "Text Generator," then select Install and Enable. This installs the latest published release (v0.8.7 as of this review) directly from inside Obsidian, with no account required for the base plugin.
Manual build from source
- How to Install:
- Clone github.com/nhaouari/obsidian-textgenerator-plugin, install dependencies with pnpm per the repository's own README, build the plugin, then copy the built files into your vault's plugins folder.
Installing the plugin itself requires no account or API key. You only need credentials once you configure an actual provider — a cloud API key for OpenAI, Anthropic, Gemini, or HuggingFace, or a local server address if you use the "Default (Custom)" provider with Ollama or LM Studio.
Text Generator Pricing
Text Generator itself is entirely free, with no paid tier, license key, or account of any kind. Its MIT license permits free use, modification, and redistribution. Any cost you incur comes only from the AI provider you connect — a local Ollama or LM Studio model costs nothing beyond your own hardware and electricity, while a cloud provider (OpenAI, Anthropic, Google Gemini, or HuggingFace) bills you directly per its own pricing.
Text Generator plugin
- Cost:
- Free, MIT-licensed, no account required
Local model via custom provider (Ollama or LM Studio)
- Cost:
- Free — runs on your own hardware, no per-token charge
Cloud provider (OpenAI, Anthropic, Gemini, HuggingFace)
- Cost:
- Billed directly by that provider, per your own API key and usage — not by Text Generator
Developer support (optional)
- Cost:
- Voluntary donations via Buy Me a Coffee, per the plugin's manifest — not required to use any feature
There is no version of Text Generator that costs money on its own. Every dollar you might spend using it goes to a separate cloud AI provider you configure, or is a voluntary donation to the developer — never a payment to the plugin itself.
Text Generator vs. Copilot for Obsidian
Text Generator and Copilot for Obsidian both bring AI into Obsidian, but they solve different problems: one inserts text from reusable templates, the other is a persistent chat agent that can read and edit your vault.
Interaction model
- Text Generator:
- Template-driven: run a saved prompt against a note or selection
- Copilot for Obsidian:
- Persistent chat interface, plus a multi-step Agent mode
Vault awareness
- Text Generator:
- Works on the active note or selection, using "Considered Context" for surrounding content
- Copilot for Obsidian:
- Searches and reads across your whole vault automatically
File editing
- Text Generator:
- Inserts generated text into the note you are working in
- Copilot for Obsidian:
- Agent mode can create and edit notes across multiple steps
License
- Text Generator:
- MIT
- Copilot for Obsidian:
- AGPL-3.0
Local model setup
- Text Generator:
- Generic "Default (Custom)" provider — manually enter a local base URL and model name
- Copilot for Obsidian:
- Named Ollama and LM Studio providers, plus a generic BYOK local slot
Pricing
- Text Generator:
- Entirely free; no paid tier at all
- Copilot for Obsidian:
- Free BYOK/local tier, plus optional paid Lite/Plus tiers for hosted models
If your workflow is "run the same kind of prompt repeatedly and drop the result into a note" — outlines, summaries, titles — Text Generator's template system fits that directly. If you want an ongoing chat that reads your whole vault and can edit files across multiple steps, see the full Copilot for Obsidian review instead.
Who Should Use Text Generator?
Whether Text Generator fits depends on whether your workflow centers on repeatable generation tasks rather than open-ended conversation.
Use Text Generator if
- You want to run the same kind of prompt repeatedly — outlines, titles, summaries — as a reusable template rather than retyping it each time
- You are comfortable manually entering a local server's base URL to connect a fully local model, rather than selecting a named "Ollama" option from a list
- You want to browse and reuse templates other users have built, instead of writing every prompt yourself
- You want an actively maintained plugin — this review found a commit as recently as August 2026 and a release in April 2026
Wait, or look elsewhere, if
- You want a persistent chat interface that references your whole vault automatically — see Copilot for Obsidian instead
- You want a one-click, named local-model provider rather than manually configuring a custom base URL and model name
- You need multi-step agent tasks that edit files across your vault — Text Generator inserts text into the note you are actively working in, not a multi-file autonomous workflow
- You need official vendor support with an SLA — Text Generator's support channels are its GitHub Issues page and community, typical of an open-source project rather than a vendor with a service agreement
Competitors and Alternatives
Text Generator competes with several other AI plugins for Obsidian and similar notes apps that add chat or text-generation capability on top of a local or cloud model.
- Copilot for Obsidian — a heavier, whole-vault AI agent plugin with a persistent chat interface, multi-step file editing, and named Ollama/LM Studio providers; see the comparison section above for a direct match-up.
- BMO Chatbot — a lightweight chat sidebar scoped to the note you have open, rather than Text Generator's template-driven insertion model.
- Obsidian and Logseq AI Plugins, Compared — a multi-plugin comparison covering Text Generator, Copilot for Obsidian, BMO Chatbot, Smart Connections, and logseq-copilot side by side.
Common Mistakes When Evaluating Text Generator
Most confusion about Text Generator comes from expecting a named local-model provider, or confusing it with an unrelated tool of a similar name.
Mistake 1: Expecting a dedicated "Ollama" or "LM Studio" option in the provider list
Text Generator's named providers are OpenAI, Anthropic, Google Gemini, and HuggingFace. Local models are reached through the generic "Default (Custom)" provider, where you manually enter a base URL and model name — there is no automatic local-server picker.
Mistake 2: Confusing Text Generator with text-generation-webui
text-generation-webui (also called "oobabooga") is a separate, standalone web UI for running local language models, with no relationship to this Obsidian plugin beyond the coincidental similarity in name.
Mistake 3: Assuming Text Generator is a chat interface
Text Generator is template-driven: you run a saved prompt against a note or selection and it inserts the result directly into your document. For a persistent, vault-aware chat interface, see the Copilot for Obsidian comparison above.
Mistake 4: Assuming stale, cached star counts
GitHub star counts change regularly. This review verified 1,985 stars directly via the GitHub API on September 12, 2026 — re-check that endpoint yourself for a current figure.
Frequently Asked Questions
What is Text Generator?
Text Generator is a free, MIT-licensed community plugin for the Obsidian notes app that inserts AI-generated text into notes using reusable, frontmatter-configured templates, built by developer Noureddine Haouari.
Is Text Generator free?
Yes. The plugin itself is entirely free under the MIT license, with no paid tier or license key. Any cost comes only from an optional cloud AI provider you configure, since local models via a custom OpenAI-compatible endpoint are free to run on your own hardware.
Does Text Generator support local AI models like Ollama?
Yes, but not through a dedicated "Ollama" option. Its generic "Default (Custom)" provider accepts any OpenAI-compatible base URL, which is the documented way to connect it to a local server run through Ollama, LM Studio, or a similar tool.
What license is Text Generator released under?
The MIT License, per the plugin's own manifest.json on GitHub.
Is Text Generator the same as text-generation-webui?
No. text-generation-webui (also called "oobabooga") is a separate, standalone web application for running local language models. It has no connection to this Obsidian plugin beyond a similar-sounding name.
Does Text Generator work like a chat interface?
No. It is template-driven — you run a saved prompt against a note or selection, and it inserts the generated text directly into that note, rather than displaying a back-and-forth conversation in a chat panel.
How many GitHub stars does Text Generator have?
Its repository (github.com/nhaouari/obsidian-textgenerator-plugin) showed 1,985 stars as of September 12, 2026, verified directly via the GitHub API.
Is Text Generator actively maintained?
Yes. This review found a commit as recently as August 2026 and a tagged release, v0.8.7, published April 27, 2026 — check the repository directly for anything more recent.
How do I install Text Generator?
In Obsidian, open Settings → Community plugins → Browse, search for "Text Generator," then select Install and Enable — no account is required for the base plugin.
Who develops Text Generator?
Developer Noureddine Haouari, with a companion site at text-gen.com and documentation at docs.text-gen.com, not affiliated with Obsidian's maker.