Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
FAIR Chemistry's library of machine learning methods for chemistry
| Date | Stars |
|---|---|
| 2026-07-31 | 2205 |
| 2026-08-06 | 2213 |
| 2026-08-07 | 2213 |
| 2026-08-08 | 2213 |
| 2026-08-09 | 2213 |
| 2026-08-10 | 2212 |
| 2026-08-11 | 2215 |
| 2026-08-12 | 2218 |
| 2026-08-13 | 2219 |
| 2026-08-14 | 2219 |
| 2026-08-15 | 2221 |
| 2026-08-17 | 2222 |
| 2026-08-18 | 2223 |
| 2026-08-19 | 2224 |
| 2026-08-20 | 2225 |
| 2026-08-21 | 2226 |
| 2026-08-22 | 2226 |
| 2026-08-23 | 2227 |
| 2026-08-24 | 2230 |
| 2026-08-25 | 2231 |
| 2026-08-27 | 2231 |
| 2026-08-29 | 2231 |
| 2026-08-31 | 2235 |
| 2026-09-01 | 2235 |
| 2026-09-02 | 2236 |
| 2026-09-03 | 2236 |
| 2026-09-04 | 2237 |
| 2026-09-06 | 2238 |
| 2026-09-07 | 2239 |
| 2026-09-08 | 2240 |
| 2026-09-09 | 2242 |
| 2026-09-11 | 2248 |
| 2026-09-12 | 2250 |
| 2026-09-13 | 2251 |
| 2026-09-14 | 2252 |
| 2026-09-15 | 2253 |
| 2026-09-16 | 2254 |
| 2026-09-18 | 2258 |
| 2026-09-19 | 2259 |
| 2026-09-20 | 2260 |
| 2026-09-21 | 2261 |
| 2026-09-22 | 2262 |
| 2026-09-23 | 2263 |
| 2026-09-24 | 2266 |
| 2026-09-25 | 2267 |
| 2026-09-26 | 2267 |
| 2026-09-27 | 2267 |
| 2026-09-29 | 2270 |
| 2026-09-30 | 2272 |
| 2026-10-01 | 2273 |
| 2026-10-02 | 2274 |
| 2026-10-03 | 2275 |
| 2026-10-05 | 2277 |
| 2026-10-06 | 2279 |
| 2026-10-07 | 2279 |
| 2026-10-08 | 2281 |
| 2026-10-09 | 2284 |
| 2026-10-10 | 2285 |
| 2026-10-12 | 2286 |
| 2026-10-13 | 2288 |
| 2026-10-14 | 2291 |
| 2026-10-15 | 2293 |
| 2026-10-16 | 2295 |
| 2026-10-18 | 2296 |
| 2026-10-19 | 2297 |
| 2026-10-20 | 2297 |
| 2026-10-21 | 2297 |
| 2026-10-22 | 2299 |
| 2026-10-23 | 2313 |
| 2026-10-24 | 2319 |
| 2026-10-25 | 2321 |
| 2026-10-26 | 2322 |
| 2026-10-27 | 2325 |
| 2026-10-28 | 2329 |
| 2026-10-29 | 2332 |
| 2026-10-30 | 2335 |
| 2026-10-31 | 2337 |
| 2026-11-01 | 2338 |
| 2026-11-02 | 2339 |
| 2026-11-03 | 2340 |
| 2026-11-04 | 2343 |
Today
+3 stars today
This week
+14 stars this week
This month
+66 stars this month
Momentum
61.0
growth rate 0.60%/day
<p align="center">
<picture>
<source media="(prefers-color-scheme: dark)" srcset="docs/assets/fair-chemistry-logo-dark-mode.png">
<source media="(prefers-color-scheme: light)" srcset="docs/assets/fair-chemistry-logo-light-mode.png">
<img src="docs/assets/fair-chemistry-logo-dark-mode.png" width="650" alt="FAIR Chemistry">
</picture>
</p>
<h4 align="center">
[](https://codecov.io/gh/facebookresearch/fairchem)
[](https://doi.org/10.5281/zenodo.15587498)
[](https://github.com/codespaces/new/facebookresearch/fairchem?quickstart=1)
</h4>
# `fairchem` by the FAIR Chemistry team
`fairchem` is the [FAIR](https://ai.meta.com/research/) Chemistry's centralized repository of all its data, models,
demos, and application efforts for materials science and quantum chemistry.
- [UMA performance](#uma-is-now-much-faster)
- [Latest news](#latest-news)
- [Installation](#installation)
- [Quick Start](#quick-start)
- [Pretrained models](#instantiate-a-calculator-from-a-pretrained-model)
- [Tasks](#set-the-task-for-your-application-and-calculate)
- [Inorganic crystal relaxation](#relax-an-inorganic-crystal)
- [Molecular dynamics](#run-molecular-dynamics-md)
- [D3 dispersion corrections](#d3-dispersion-corrections)
- [Multi-GPU inference and LAMMPS](#multi-gpu-inference-and-lammps)
- [License](#license)
## UMA is now much faster!
<p align="center">
<a href="benchmarks/omol/README.md">
<img src="benchmarks/omol/omol_force_mae_vs_speed.png" width="650" alt="OMol force MAE versus ASE NVE runtime">
</a>
</p>
Uma-nitro brings blazing speedups to UMA. Currently it is only supported on Nvidia Ampere GPUs and above. Speed is heavily optimized on Hopper series and above for single-gpu single-system mode. On unsupported GPU types, the system will fallback to other non-nitro modes. Stay tuned as we bring nitro support for multi-gpu and batch modes. In the sub 5-10 ms/step regime, ASE integrators starts to become a bottleneck. To achieve the true performance capabilities of these models, we recommend moving to non-python/GPU based integrators.
The vertical axis is force MAE (meV/Å) on the public OMol25 validation set; the
horizontal axis is MD speed (ns/day at a 1 fs timestep) from the mean runtime of an ASE
NVE step on a 1000-atom silicon cell using an NVIDIA H200. See the
[benchmark details and reproduction scripts](benchmarks/omol/README.md)
for the full setup.
For more OMol model accuracy benchmarks, you can also see the
[FAIR Chemistry leaderboard](https://huggingface.co/spaces/facebook/fairchem_leaderboard).
Please use this logo for references to [UMA](docs/assets/uma-diagram-light-mode.png)!
## Latest news
### October 2026
- **UMA Nitro**, the new default inference mode, is much faster than turbo for molecular dynamics and relaxations on a single GPU.
On a single NVIDIA H200, it reaches **~6 ns/day for 1,000 atoms** in ASE MD
and raw inference throughput equivalent to **~33 ns/day for 150 atoms**,
with support for systems up to **15,000 atoms**.
Special thanks to Emanuele Arcelli and Max Zhdanov (for their [FlashUMA work](https://github.com/ArcelliEmanuele/flash-uma)), Carole-Jean Wu, Christian Puhrsch, Zachary DeVito, Aditya Venkataraman, Michael Lazos, and Animesh Jain, all of whom contributed to and helped improve UMA Nitro.
See [UMA Nitro](https://facebookresearch.github.io/fairchem/ase-calculator#default-mode-nitro).
- **Two new UMA-S-1.2.1 models**: `uma-s-1p2p1-ft-d4rev` for molecules (ωB97M-D4rev) and
`uma-s-1p2p1-ft-r2scan` for materials (Excerpt of 13,900 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:768036fbad8fc00e, llm:Repository description: 'FAIR Chemistry's library of machine learning methods for chemistry' (facebookresearch/fairchem).
matched fp:768036fbad8fc00e, llm:Repository description: 'FAIR Chemistry's library of machine learning methods for chemistry' (facebookresearch/fairchem).
matched fp:768036fbad8fc00e, llm:Repository description: 'FAIR Chemistry's library of machine learning methods for chemistry' (facebookresearch/fairchem).
