![ERROR on proc 3: Insufficient memory on accelerator" when running GPU using lammps_v29Sep21 - LAMMPS Installation - Materials Science Community Discourse ERROR on proc 3: Insufficient memory on accelerator" when running GPU using lammps_v29Sep21 - LAMMPS Installation - Materials Science Community Discourse](https://matsci.org/uploads/default/original/2X/9/9ed47ec2a7b401e49a6db461a4f60ddbc1337aec.png)
ERROR on proc 3: Insufficient memory on accelerator" when running GPU using lammps_v29Sep21 - LAMMPS Installation - Materials Science Community Discourse
BUG] Using pair style hybrid with GPU package and neighbor lists on GPU · Issue #2621 · lammps/lammps · GitHub
![NVIDIA Claims Ampere A100 Offers Up To 2x Higher Performance & 2.8x Efficiency Versus AMD Instinct MI250 GPUs NVIDIA Claims Ampere A100 Offers Up To 2x Higher Performance & 2.8x Efficiency Versus AMD Instinct MI250 GPUs](https://cdn.wccftech.com/wp-content/uploads/2022/06/NVIDIA-A100-delivers-up-to-1.9X-higher-performance-when-running-popular-HPC-applications-on-a-single-GPU.png)
NVIDIA Claims Ampere A100 Offers Up To 2x Higher Performance & 2.8x Efficiency Versus AMD Instinct MI250 GPUs
![Model Particles in Liquid, Solid, or Gaseous States Using LAMMPS Certified GPU Systems | Exxact Blog Model Particles in Liquid, Solid, or Gaseous States Using LAMMPS Certified GPU Systems | Exxact Blog](https://images.contentstack.io/v3/assets/blt71da4c740e00faaa/bltaed722f281fb0faa/602afc8264f941792ae6d6d2/lammps-Chart.png)