Claude Fable 5.2 autonomously speeds up Quantum Espresso DFT simulations by 36x
simonbatzner · x · 2026-11-10
In early testing, discoveredMat used Claude Fable 5.2 to speed up Quantum Espresso's DFT and DFPT codepaths by 36x and 9x respectively — with the model running largely in an autonomous loop, checking its own work against verification sets.
Quantum Espresso is the most widely used open-source DFT codebase (600,000 lines of Fortran), making an autonomous agent finding and validating optimizations in it a notable scientific-computing result.
Related event: AI Agent Rewrites Quantum Code for 36x Speedup(2 posts)→
More from Models
- Anthropic gives loyal Claude Code subscribers up to 3 Fable resets based on tenure — threepointone · 2026-11-10
- Ant Group's Robbyant open-sources LingBot Touch, a JEPA-based tactile foundation model — AdinaYakup · 2026-11-10
- Fable 5.2 not yet showing up in Claude Code for some users — DavidmComfort · 2026-11-10
- Claude Fable 5.2 designs 3 working electronic devices from one-paragraph prompts — Dr_Singularity · 2026-11-10
- Salesforce's Interactive Visual Gym: best VLM finishes only 53.4% of 315 closed-loop tasks — jcniebles · 2026-11-10
- Benchmark audit: 5 of 6 AI benchmarks have grading defects; only PostTrainBench verified — maksym_andr · 2026-11-10