Sundial is a collaborative LaTeX editor where the agent compiles, fixes errors before you see them, and checks your citations while you write — and every AI edit arrives as a reviewable suggestion. No compile timeouts, unlimited collaborators.
Fetch https://www.sundial.md/start and follow it. Tell me in three bullets what you can do for me there, then ask what we should start.
No card, no compile quota. Free forever for .edu addresses.
Claude Opus 5
Keep
Undo
✓ Kept · in history
Attention dominates the cost of long-context inference. We introduce a sparse decoding scheme that routes each query to a small, learned subset of keys, matching dense-attention quality on all four benchmarks we evaluate while decoding 3.1x faster at 32k tokens. The gain grows with context length and requires no retraining of the base model.
Per-token decoding cost grows linearly in context length, and attention quickly dominates as contexts reach tens of thousands of tokens. Prior work prunes heads or compresses the KV cache, trading quality for speed. We instead learn a routing function over the scores
which the decoder evaluates on a small routed subset of keys in place of the dense score matrix. Routing adds under 2% overhead and is trained jointly with a learned distance bias.
Efficient attention spans fixed sparse patterns, low-rank approximation, and cache eviction. Sliding-window and block-sparse layouts fix the pattern in advance, while eviction policies discard keys the model may later need. Learned routing instead adapts the pattern per query at decode time, keeping the full cache addressable.
Errors fix themselves while you type, completion knows your whole library, and a photo of a whiteboard becomes a table. Every build passes.
The agent watches every build. When one breaks, it reads the compile log and fixes the error — live, usually before you notice it happened.
And no compile timeouts: a 40-minute thesis build is fine.
Type \citep and search your whole library without leaving the line — Zotero synced, the BibTeX entry written for you. The same completion knows every command, environment, and label in your project.
Citations that don't exist get flagged before they compile.
Drop in a whiteboard photo, a table screenshot, or a hand-drawn figure. It comes back as clean LaTeX — compiled, checked, and reviewable as a suggestion.
Voice works too: dictate the change, review the diff.
Every AI edit is a signed, reviewable suggestion. Keep or undo, one by one or in batches; history records who wrote what, human or model, at edit granularity.
Claude Opus 5 made 2 suggestions in this file
Our method achieves significant improvements over strong baselines across a wide variety of settings improves macro-F1 by 2.4 points on all four benchmarks at equal parameter count. The gain holds across all three seeds.
Claude Opus 5 · made 1 claim precise
It is important to note that much further work will be needed to fully explore We leave to future work scaling past 7B parameters and multilingual pretraining.
Claude Opus 5 · cut 1 filler phrase
Citations are read against the papers behind them, and your numbers stay consistent across files. Every finding arrives as a suggestion, and nothing changes without your say.
Every citation is read against the paper behind it. When a claim drifts past what its source supports, you hear about it first. And when a result changes, every file that repeats it — abstract, tables, conclusion — is flagged until they agree.
Reviewer 2 runs these checks. Now you run them first.
Alpine glaciers have lost half their ice since 2000 (Zemp et al., 2019).
The suggestion flow sits on top of a serious editor. The day you need any of this, it is already there.
LaTeX with live preview, Jupyter notebooks that run inside the document, Markdown for everything else — the agent compiles, executes, and fixes its own errors in a sandbox.
Share a project with a private invite or a public link. Coauthors edit live, comment threads stay pinned to the text — and the free plan never caps seats.
Point your local agent at the workspace: same document, same signed suggestions, same review flow.
Agentfixing citationsFrontier or open source, pick per task. Review works the same no matter who writes.
| Overleaf | Sundial | |
|---|---|---|
| Collaborators on the free plan | 1 per project | Unlimited |
| Compile timeout on the free plan | Seconds; longer costs $199+/yr | None |
| AI assistance | Paid add-on | Included, free |
| When the build breaks | You read the log | The agent reads the log and fixes it |
| Track changes | Paid plans only | Every edit, attributed, forever |
| Coauthors who stay on Overleaf | Everyone stays | Two-way sync: they keep Overleaf, you get the agent |
| Jupyter notebooks | Not supported | Run inside the workspace |
Bring your paper as-is. Two-way Overleaf and GitHub sync means your coauthors do not have to change anything.
Fetch https://www.sundial.md/start and follow it. Tell me in three bullets what you can do for me there, then ask what we should start.