AlphaFold Models Have Fewer Parameters but Higher Compute Costs Than LLMs
Gabriele Corso · 🔬Generating Molecules, Not Just Models · Feb 12, 2026 · at 35:06
Gabriele Corso, co-founder of Boltz, explains why structural biology foundation models scale differently than text-based large language models.
0:00 / 0:31exact quote · 31.1s
▶ Watch the full episode on YouTube → 720p mp4 · rendered on demand · StarZero watermark
“They, in terms of parameters, are actually not very big. They are definitely below a billion parameters. You know, if you're here these days in LLM space, you know, a model with less than a billion parameters, you'd think can't do anything. But on the other hand, when you look at the computational cost of running these models, they are actually a lot more expensive than it is to run a language models, because as Jeremy was saying, we go from instead of having sort of like quadratic operations, In our cubic operation.”
quote is from the automated transcript, cleaned for reading: filler sounds and stutters are removed, nothing is rephrased. names can be misheard (the analysis reads context, assessments check outside sources). how →
More from Gabriele Corso
Specialized Equivariant Architectures Vastly Outperform Simple Transformers in Molecular ML
“This field is one of the I would argue very few fields in applied machine learning where we still have kind of architecture. They are Very specialized. And, you know, there are many people that have tried to replace these architectures with, you know, simple t…”
Boltz-1 Was the First Open-Source Model Matching AlphaFold 3 Accuracy
“We went ahead and built Pulse One, which was the first fully open source kind of model to approach the level of accuracy of half-fold-free.”
Generative Molecular Models Exhibit Dramatic Inference-Time Compute Scaling
“These days we're seeing kind of pretty dramatic inference time scaling of these models where, you know, the more you run them, the better the results are.”
Model Structural Confidence Is a Poor Predictor of Binding Affinity
“Unfortunately, confidence is not a very good predictor of affinity.”
Most Protein Design Validation Uses Targets With Training Data Overlap
“One of the things that, you know, we found, ah, with the field was that a lot of the validation, especially outside of the validation that was done on specific problems, was done on targets that have a lot of, you know, known interactions in, in the training d…”