# Lune vs Undermind

Undermind spends one long search reading hundreds of candidates. Lune breaks the same work into short calls the agent chains itself. Both ship real MCP servers, so the differences are underneath.

Competitor site: https://undermind.ai.

## Choose Undermind if

- You want one exhaustive sweep, preprints and biomedical journals included.
- Your topic sits outside computer science.
- You want paper libraries and folders living with the search tool.

## Choose Lune if

- You want the agent to pick its next move, not wait on one verdict.
- You need claims tied to a quoted sentence, not a reading list.
- You compare methods across papers and want a table for it.
- You want reviewer expectations and ablation guidance in the same connection.

## From a question to a citation

Undermind: One deep search -> Read hundreds of papers -> Ranked reading list.
Lune: Chain short calls -> Read the full text -> Quote with a locator.

One long search, or a set of moves the agent chooses between.

## Line by line

| Dimension | Lune | Undermind |
| --- | --- | --- |
| The core motion | A dozen small tools: search, sweep, full text, citations, verify. | One deep search, plus paper and workspace tools around it. |
| What is indexed | Top-tier venues, most papers in full text. | Semantic Scholar: 200M+ articles, arXiv and PubMed included. |
| What is left out | Everything outside a CORE or CCF top-tier venue. | Little, by design. Ranking is left to the search. |
| How deep a citation goes | A verbatim quote and its locator, per claim. | Inline citations, with full-text PDFs read in parallel. |
| Time to a useful answer | Seconds, so the agent can branch on what it finds. | Minutes. That is what makes it thorough. |
| Entry price | $4.99 a month, or $47.90 a year. | $16 a month, billed annually. |
| Methodology guidance | A corpus on peer review, ablations and venue choice. | None published. |

## One long search, or a set of moves

Undermind bets that literature search is one careful, expensive operation. Give it a goal and it works through hundreds of candidates, then tells you how confident it is that nothing was missed. On a genuinely open question that is hard to beat.

Lune bets that an agent mid-task rarely has one clean question. It has a paragraph it is drafting and a number it half remembers. So the tools are small: one question, twenty-five angles, the citation graph, a typed table, a claim check. The agent picks the next call from the last result.

## Coverage against selectivity

Undermind searches 200 million records. Lune reads what cleared review at a venue the field respects. On coverage this is not a contest.

It is a contest on what comes back. Everything Lune returns already passed peer review, so the model spends no judgment deciding whether a source counts. The cost is real: an arXiv-first subfield will have work we do not hold.

## Both ship MCP, so look at the tools

Undermind's MCP server is thorough, free-tier, and works in every client worth naming. Anyone selling you agent access to the literature as a Lune-only capability is selling you something.

The shape differs. Much of Undermind's toolset manages its own workspace: folders, starred papers, saved searches. Lune's are evidence operations, and there is nothing to file afterwards because the output lands in your draft.

## 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
- Undermind Free: $0, standard limits, agents included
- Undermind Pro: $16, a month annually, 10x limits
- Undermind Team: $15, per person a month, annually
- Undermind Enterprise: Custom, SSO, SLA, support

Undermind publishes annual rates; monthly may differ.

## Questions

### Is Lune an Undermind alternative?

For day-to-day work inside an agent, yes. For one exhaustive sweep of everything published, Undermind covers ground Lune does not hold.

### Does Undermind have an MCP server?

Yes, at mcp.undermind.ai, and its free plan includes connecting agents. Both products are agent-native.

### Which is faster?

Lune, for one question. A search returns in seconds so the agent can chain the next call. A deep search is deliberately slow.

### Does Lune cover preprints?

Only where the paper was accepted at an indexed venue. If it lives on arXiv and nowhere else, it is not in the corpus.

## What we read

- [undermind.ai](https://undermind.ai)
- [Pricing](https://undermind.ai/pricing)
- [MCP server](https://undermind.ai/mcp)