# Lune vs ScholarAI

ScholarAI was early to putting real papers inside ChatGPT, and its reach shows: a GPT, an API, an MCP server, and drafting in Jenni. Lune covers less of the literature and goes deeper into it.

Competitor site: https://scholarai.io.

## Choose ScholarAI if

- ChatGPT is where you work and you want a GPT that already searches papers.
- You want drafting and auto-citation in the same product, with Zotero sync.
- Your topic spans clinical or life-science journals.

## Choose Lune if

- You move between clients and want one connection that behaves the same.
- You need the exact sentence and its locator, not a link.
- You compare methods across papers and want one typed table.
- You want reviewer expectations in the same connection as the papers.

## From a question to a citation

ScholarAI: Search from ChatGPT -> Save to a library -> Auto-cite in Jenni.
Lune: Search from any agent -> Read the full text -> Quote with a locator.

Collecting sources and checking a sentence are different jobs.

## Line by line

| Dimension | Lune | ScholarAI |
| --- | --- | --- |
| Where it runs | Any MCP client, plus a CLI and three miniapps. | A ChatGPT GPT, an API, an MCP server, and Jenni. |
| What is indexed | Top-tier venues, most papers in full text. | Millions of peer-reviewed papers across disciplines. |
| How deep a citation goes | A verbatim quote and its locator, per claim. | Papers summarized and auto-cited into the draft. |
| Comparing several papers | One typed table across the papers you name. | Agentic chat over your library. |
| Writing and citation styles | None. Your agent writes; Lune checks. | Drafting with auto-citation in Jenni. |
| Entry price | $4.99 a month, or $47.90 a year. | Jenni Plus at $12, Pro at $29. |

## Collecting sources, or checking sentences

ScholarAI is good at the first. Ask in ChatGPT, get real papers, save them, sync to Zotero, let Jenni drop them into the draft. That is a short path from question to bibliography.

Lune is built for what happens after the draft exists. verify_claims answers each sentence with the quote that supports it, or with nothing at all. The point is not a faster bibliography. It is that reference 14 says what you said it says.

## One connection that outlives your client

ScholarAI reaches most people where they already are, and has done for longer than most. Its writing path runs through Jenni, so the further you get from ChatGPT and Jenni, the more of the product you leave behind.

Lune connects one way, through the protocol. The same tools answer in Claude Code, Cursor, Codex and ChatGPT, with the same scopes and quota. Switch editors and nothing about your setup changes.

## Pricing

- Lune Free: $0, 10 requests a day, every tool
- Lune Pro: $4.99, a month or $47.90 a year, 300 requests a day
- Lune Max: $9.99, a month or $95.90 a year, 600 requests a day
- Lune Credits: $0.99, 300 extra requests, no expiry
- ScholarAI Free: $0, limited chat and uploads
- ScholarAI Plus: $12, a month, 5,000 autocompletes
- ScholarAI Pro: $29, a month, unlimited drafting

ScholarAI's drafting happens in Jenni, so these are Jenni's plans.

## Questions

### Is Lune a ScholarAI alternative?

For grounding an agent in top-tier computer science literature, yes. For drafting with auto-citation inside ChatGPT and Jenni, no.

### Does ScholarAI have an MCP server?

Yes. ScholarAI published one aimed at Claude Desktop and other agent systems, with scoped API keys.

### Can Lune cite in a specific style?

Lune returns metadata and quotes; your agent formats the citation. There is no built-in style engine.

## What we read

- [scholarai.io](https://scholarai.io)
- [MCP announcement](https://scholarai.io/blog/mcp)
- [Jenni pricing](https://jenni.ai/pricing)