Users are praising Boltz's new SOTA models for protein and small molecule design because early API tests produced remarkable results and impressive performance.
No Digg Deeper questions have been answered for this story yet.
BoltzMol-1: Our first small-molecule hit-discovery pipeline leveraging an optimized affinity model and a new suite of ADMET models. We tested BoltzMol on 10 very challenging OOD targets and we got confirmed hits on 6, while only testing a few compounds per target (28–51).
@AnthropicAI @benchling @phylo_bio @amazon @tamarindbio @PaulingAi @mirrorphysics @CultivariumFRX To celebrate, we're giving $2,000 in API credits to every company and $100 to every academic for the first two weeks. Grab a key at http://api.boltz.bio and start designing today!
We’ve partnered with @AnthropicAI, @benchling, @phylo_bio, @amazon Bio Discovery, Rowan, @tamarindbio, Kiin Bio, @PaulingAi, @mirrorphysics and @CultivariumFRX on first-party integrations so that our newest models and primitives are natively supported in your favorite tool.
BoltzProt-1: Our most powerful protein design pipeline, powered by BoltzPPI, a new protein-protein affinity model that prioritizes designs most likely to bind. On hard targets, BoltzProt nearly tripled BoltzGen’s VHH hit rate and outperformed other leading proprietary models, while 58% of binders passed a stringent developability panel, matching clinical-stage therapeutic nanobodies.
To our knowledge, this is the most extensive experimental validation of AI-driven small molecule hit discovery to date. With BoltzMol, we can now go from target to validated hits in 3 to 8 weeks for under $15k, making hit discovery dramatically faster, cheaper, and more accessible than traditional HTS.
@AnthropicAI @benchling @phylo_bio @amazon @tamarindbio @PaulingAi @mirrorphysics @CultivariumFRX More details on the models on: - our blog post: https://boltz.bio/boltzmol-boltzprot-api - BoltzMol-1 manuscript: https://boltz.bio/boltzmol1-technical-report.pdf - BoltzProt-1 manuscript: https://boltz.bio/boltzprot1-technical-report.pdf
@AnthropicAI @benchling @phylo_bio @amazon @tamarindbio @PaulingAi @mirrorphysics @CultivariumFRX Join us for our launch events: - Boston, June 18th: https://luma.com/fpk6ls8j - San Francisco, June 22nd: https://luma.com/3tid086q - London, July 2nd: https://luma.com/dv5jgjaa
The Boltz API: The fastest, cheapest and most flexible way to run Boltz models from just $0.025 per prediction on high-performance distributed compute. Built from the ground up for agentic workflows, the API is a powerful new set of primitives for drug discovery and biomolecular design.
Don’t get fooled: these designs are still far from zero-shot therapeutic biologics. BoltzProt-1 does not remove the need for downstream optimization, but it can give you a meaningful head start by reducing the time and cost required to reach strong starting points for your therapeutic program.
@AnthropicAI @benchling @phylo_bio @amazon @tamarindbio @PaulingAi @mirrorphysics @CultivariumFRX We’ve also built support for OpenAI Codex and Gemini CLI so you don’t have to choose. As always: you own your inputs and outputs, and we don't train on your data.
@GabriCorso Been running ternary predictions and cofloding de-novo predictions through the Boltz API, and the results have been remarkable. The agentic workflow primitives are a game changer. The experimental validation numbers are impressive. Keep pushing!
@GabriCorso @AnthropicAI @benchling @phylo_bio @amazon @tamarindbio @PaulingAi @mirrorphysics @CultivariumFRX Amazing work! Will you publish model and code on the GitHub? I would love to analyze it and benchmark on proprietary data, so I can’t use api
@GabriCorso @RShprints Any benchmarks?
@GabriCorso Thank you!
@GabriCorso amazing
@GabriCorso Congrats!
@GabriCorso 👀
@GregPreibisch @AnthropicAI @benchling @phylo_bio @amazon @tamarindbio @PaulingAi @mirrorphysics @CultivariumFRX Hi Greg, no these set of models won't be open source. However our API is secure and we offer also on premise deployment options
@GabriCorso @AnthropicAI @benchling @phylo_bio @amazon @tamarindbio @PaulingAi @mirrorphysics @CultivariumFRX Love that the quick start has recommended prompts for agents!
@GabriCorso Amazing update, Gabriele!