leonickson

no Anthropic involvement, I just used Claude Code heavily while building this and putting that in the README felt more honest than not mentioning it. Now that I think about it may be it shuld be "built with claude code" instead of "built in collaboration...". Changed it.

It's tunable, --cache-gb N on the CLI. In my sweep the speed barely moved between a 1GB and 6GB cache (43% vs 70% hit rate, same tok/s) because right now the bottleneck is GPU dispatch, not the SSD. so more RAM doesnt buy much yet. once the kernel work lands it should start to matter, so on 32GB I would just set 8 and let it age well. Also the hit rates themselves answer the "can you even know which experts stay hot" question, reuse across tokens is very real.

agree, prefill is the weak spot right now. it goes through the same per-token path as decode, which is dumb for long prompts. The fix is on the list: during prefill we can batch the expert reads for the whole prompt per layer instead of per token, that amortizes the IO a lot. until that lands, honest answer is this is good for chat-length stuff, not for feeding it a 10k token document.

Thanks

It is free, for higher limits you have to sign up. Thanks.

I have updated the terms, and all the community submissions are under MIT.

Thanks! From what I've seen, LLMs confidently generate SwiftUI that looks off from what you want on the first try, then you're spending a couple of rounds getting it to fix it. I am just making it easier. These are all real, compiled, and Simulator-rendered, so you start from something that actually builds. The MCP just hands them to your agent so it composes from working components instead of guessing.

Also, this is community based, if you have some awesome components and would be happy to share. Please do so. Thanks

The extension is live in Firefox - https://addons.mozilla.org/addon/tomesphere/

Yes, DCR is supported (WorkOS exposes the registration endpoint). What error is Aperture throwing? I suspect a metadata-discovery detail.

I really like your work with the gradients.

Thank you so much! I'd genuinely like to hear more, you can email me here - help@tomesphere.com

Hello,

1. Yes, I am working on it, I will soon release it for Firefox and Safari (by this weekend). 2. For now I am just trying to see what people want, may be the business model is more focused on mcp for higher limits for labs, lets see where is goes. 3. The paper/article search already has author search, I can add the university search soon in both the paper/article and altas/map search.

Thank you for all the questions and suggestions.

I am sorry, it should have taken you to the paper page/article, I will investigate it, but I did the same search and clicked on the first link, it to me to this - https://tomesphere.com/paper/2301.01365

Thanks for letting me know.

Will check it, it will be really useful. Thanks

It is all open, bulk-downloadable sources. Here are some links.

  - arXiv: bulk on S3 (https://info.arxiv.org/help/bulk_data_s3.html)
  - PubMed Central Open Access: AWS Open Data (s3://pmc-oa-opendata).
  - bioRxiv / medRxiv: their monthly S3 dumps
  - OpenAlex: for citations, metadata, abstracts (https://developers.openalex.org/download/download-to-machine) 
  - AlphaFold — structures (https://alphafold.ebi.ac.uk/download)
  

Thanks for the question.

The project has four parts, and I think you may have used the search in the navigation bar. That search is for the paper page (detailed information about that specific paper). When you search, you get a list of results, and if you click one of them, it takes you to a paper page with all the in-depth details about that paper. I should have made the map UX better.

To highlight things in the atlas or map, you might want to go to the filter panel on the left side and scroll down a bit. You will see a search area that helps you search for genes, diseases, and proteins. However it might not highlight any dot for the “ribosome” because filter search for now is only connected to genes, diseases and protein. I noted this, and I will improve it. I may also move that filter search to a different place. Thank you so much.

Tomesphere includes web pages and a browser extension overlays all of this directly on the arXiv, PMC, bioRxiv, Google Scholar, and medRxiv pages you are already reading. I had noticed the Paperclip MCP before, and from what I can see, they have very good data. In some cases, they may even have better data. We also have some additional sources, such as peer review from OpenReview, video links from YouTube and SlidesLive, GitHub links, AlphaFold protein entities, citations, and semantic neighbors.

Thank you for the questions. Will improve the project more.

Hi, I love the genre too. Cosmograph is wonderful, I did try it, but because of its license restriction I could not use it for this project. I do agree that beyond an initial "that's cool" this map may not contribute much, "and that's why I didn't make it the main product. I already had the data as I was building other things (extension, paper page) and wanted a bit of a cool factor so people would take a look at the project. The value is what's under each dot, the enriched page (TLDR, genes/drugs/diseases, trials, 3D structures, code, datasets, full text), extension and the MCP for agents.

Hello, I agree with you, viz are just cool and might not really have a usecase. In this project map is not the product, it is 1 of 4 parts and to be honest the least important. The value is what is under each dot, the enriched page (TLDR, genes/drugs/diseases, trials, protein structures, code, datasets, full text, images, reviews, etc) and the MCP for agents. You are welcome to use whichever part of the project is most useful to you (whether that is the map, paper pages, browser extension, or MCP).