An internal Anthropic research model, supervised by two people with no background in optimisation, optimised the open models for protein structure and design. Plus a new mode that predicts systems over 10,000 tokens on a single server.
- Per Anthropic, the speed-up is about 4 times on average with minimal loss of precision, and nearly 2 times with identical outputs.
- Protein binder design with comparable in silico scores for about $150 in total.
- A competition with Adaptyv Bio: wet lab validation of over 5,000 designs.
Two people at Anthropic understand proteins but have never written code that squeezes a graphics card. In under four weeks, they supervised a model that sped up more than 30 tools widely used by molecular biologists.
That is how the company describes it, and the detail about the two of them is more interesting than the number 4.
The two numbers worth remembering
The first is $150. Their previous protein design campaign allowed up to $10,000 per target. Now a comparable in silico result costs about as much as dinner for four, and if it holds up in the lab, this kind of work moves from big institutes down to small labs.
The second is the honest admission about the giant structures. They can now be run on one machine, but they come out wrong, most likely because the models do not generalise that far beyond what they were trained on. It is good that they say so themselves, with pictures of the failures.
That is why the competition matters more than the blog. Five thousand designs will go through a real lab, and there we will see how many of the cheap in silico results become a protein that binds where it should.