I finished my estimate on required compute to make an atomic-resolution virtual cell: 10^38 FLOPs to simulate a human cell for 1 day. We should be able to do this simulation in 2074 using 200 TW of power. 1/3

Our ether0 paper was accepted at NeurIPS 2025! Very proud of the FutureHouse team!

Google scholar has a full-text index of nearly all research papers. You can use it to get counts for arbitrary phrases. I've been using this to measure popularity of things in science. For example, here's the popularity of Greek letters used in equations 1/3

I've written up some thoughts on publishing for machines. 10M research papers are published per year and there are 227M total - machines will be primary producers and readers of publications going forward. Humans can simply not keep up. It's time to think about revising the scientific paper.

HLE has recently become the benchmark to beat for frontier agents. We at FutureHouse took a closer look at the chem and bio questions and found about 30% of them are likely invalid based on our analysis and third-party PhD evaluations. 1/7

1/4 🚀 Announcing the 2025 Protein Engineering Tournament. This year’s challenge: design PETase enzymes, which degrade the type of plastic in bottles. Can AI-guided protein design help solve the climate crisis? Let’s find out! ⬇️ #AIforBiology #ClimateTech #ProteinEngineering #OpenScience

I have written up a 3.5k word/10 figure essay on how to write a reward function while avoiding reward hacking for chemistry. It covers all the ridiculous ways we had to avoid reward hacking for training ether0, our scientific reasoning model. diffuse.one/p/m1-000

FutureHouse's goal has been to automate scientific discovery. Now we used our agents to make a genuine discovery – a potential new treatment for one kind of blindness (dAMD). We had multiple cycles of hypotheses, experiments, and data analysis – including identify the mechanism.

We shipped multi-agents today! Our chemistry design agent can now call Crow, our scholarly research agents, to bring in data from literature/clinical trials/open targets while designing molecules. platform.futurehouse.org

Integrating @opentargets.org is so helpful to provide evidence for disease mechanisms independent of the literature. Here's a demo of synthesizing 78 papers and open targets to propose two novel targets for triple negative breast cancer See the answer: platform.futurehouse.org/trajectories...

We have an API for clinical trials on our platform - which means you can ask questions like "what trials will read out in June for NSCLC and how likely would you rate their success based on previous trials in the area." Pretty cool. Answer: platform.futurehouse.org/trajectories...

Here's a command that converts a DOI to bibtex:

I always look forward to FutureHouse releases. I had to do a little digging for API information so here it is for those who are interested. futurehouse.gitbook.io/futurehouse-...

We have gotten some really good responses to science questions from platform.futurehouse.org already. Both from "Crow" (short answers) and "Falcon" (deep research). It looks like this is state of the art right now!

Really happy to have this available on an API and free, today!

The plan at FutureHouse has been to build scientific agents for discoveries. We’ve spent the last year researching the best way to make agents. We’ve made a ton of progress and now we’ve engineered them to be used at scale, by anyone. Free and on API.

Sam Cox and I are giving the MIA seminar at the Broad Institute in Boston tomorrow. Going to tease some new results on something unrelated to scientific agents and squarely in domain of chemistry.

It's ridiculous, but there hasn't existed a one-liner to quickly get functional groups of a molecule. Little Friday night coding exercise to get this working. Enjoy - and let me know of any missing functional groups! I could only do a few hundred.

Half of an AI scientist is rejecting or accepting hypotheses. FutureHouse and Science Machines just put out ~300 novel hypotheses from ~50 published papers along with ground-truth data. Humans take 4.2 hours to solve these and frontier models get 10-20% correct. This is like SWE-bench for comp bio

We should start using SI notation for token counts - like 1 megatoken context window or 64 kilotoken reasoning model. Then we can write: 64kt or 1mk etc. Or you can say - "my prompt is 1.6 kilotokens" - which sounds badass

PaperQA2 can now work with clinical trials. It considers both research papers and clinical trials jointly to answer complex questions. It uses the the clinicial trials dot gov API - so it can do complex queries too. Checkout the tutorial below: futurehouse.gitbook.io/futurehouse-...

It's been about a month since the first batch of reasoning models was released. There’s been about a dozen reproductions since then and some patterns are emerging. I’ve written up my own notes on training recipes, frameworks, rumors, and major open questions. diffuse.one/p/d2-000

Image duplication has been a powerful signal for detecting scientific fraud, but is irrelevant in many fields. I've been working a bit on finding new signals like it that work across fields. I've found one using LLMs that can predict retractions, weakly, for $1 per paper. 1/4

Molecular dynamics requires a lot of expert knowledge to set-up and analyze simulations. We set out to automate it with LLM agents: MDCrow!

Has anyone actually seen a good table of contents figures (TOCs) that made them read a paper?

Humanity's last exam progress. Looking forward to humanity's last last exam_final

I'm very impressed with Operator. I've used a lot of web agents before and operator actually can function after dozens of steps, whereas most just die after 4-5. I asked it to find a new way to treat PCOS and it spent 12 minutes (~50 steps) on it.

I've been thinking about how reasoning models will change AI applied to science. The recent papers from Deepseek/AI2/MoonShotAI are showing that we can exceed humans on reasoning tasks and I've written up some reflections on the consequences diffuse.one/p/d1-007

Since we first released our RAG agent PaperQA, we've seen steady improvements to match human performance, and we now exceed expert scientists performance by ~25 points on doing literature research.

Trying some new stuff out at FutureHouse

In an era of hallucinating LLMs that are frankly ‘mostly’ useless for research, I must highlight the amazing value provided by Hasanyone.com. @andrew.diffuse.one this is a huge hit! Most useful AI tool for lit research I have found so far.

A holiday project: interactive LigandMPNN. Click the residue and adjust the temperature. Not quite ready for primetime but promising.

Finishing 2024 with one more research result! We’ve trained small language agents to do hard sci tasks: engineering proteins, manipulating DNA, and working with sci literature in a new library called Aviary. We beat humans and frontier LLMs on these tasks!

Looking for a post-doc? Come work with us + one of the amazing advisors from top university labs. Deadline soon - in February!

What if Eroom's law, the decreasing productivity of drug discovery, is a result of aggressive accounting whereby pharma companies consider as much stuff as possible as R&D to avoid taxes? I bet eroom's law holds in movies too - where losses are built to equal move revenue

Thanks to Nathan Labenz for hosting me on his podcast! We chatted about automating intellectual tasks of science, paperqa, FutureHouse moonshot, drug discovery, etc. I don't really go on podcasts or give talks much, so I talked a lot. www.youtube.com/watch?v=umFb...

🧪 E11 Bio is excited to share a major step towards brain mapping at 100x lower cost, making whole-brain connectomics at human & mouse scale feasible (🧠→🔬→💻). Critical for curing brain disorders, building human-like AI systems, and even simulating human brains. Read more: e11.bio/news/roadmap

Percentages are always thrown around to make something more "scientific." I've always wondered if you can infer something about population size from the value - like if a study reports a percentage of 66%, does that imply a larger population size than one that reports 67%? 1/5

The frequency of sample sizes in published research is not uniform. Here are fraction of sample sizes across research papers, based on google scholar searches. The most popular sample sizes are 10,50, and 100. 79 is the least popular sample size.

Here's a little drawing of some of the 1400 compounds released in @evebio.bsky.social's first data dump.

In case you were wondering, turkeys can fly like other birds but are not able to use tools like the more intelligent corvids citation: hasanyone.com?id=3db0a837

Here's a comparison of various chemical biology objects from gold nanoparticles to small molecule drugs, made with Blender.

We added another 100 citations (total 577) in the latest version of our review on LLMs and agents in chemistry. Take a look! arxiv.org/abs/2407.01603

Since creating an API that can deliver crow facts via REST, I've delivered 3,250 crow facts as JSON. Get yours today: curl facts.drugcrow.ai explanation: diffuse.one/p/d1-004

Ok got custom domain handle - ready to post