Isaac Gym - Training Ant ( nvrtc: error: invalid value for --gpu-architecture (-arch))

I download Isaac Gym from GitHub - NVIDIA-Omniverse/OmniIsaacGymEnvs: Reinforcement Learning Environments for Omniverse Isaac Gym, successfully run Cartpole example but can not run Ant. I have tried my best but could not solve the problem. Here are the infomation:

Error executing job with overrides: [‘task=Ant’]
Traceback (most recent call last):
File “scripts/”, line 201, in parse_hydra_configs
File “/home/lzy/.local/share/ov/pkg/isaac_sim-2022.1.1/exts/omni.isaac.gym/omni/isaac/gym/vec_env/”, line 195, in run
File “/home/lzy/.local/share/ov/pkg/isaac_sim-2022.1.1/exts/omni.isaac.core/omni/isaac/core/world/”, line 285, in reset
File “/home/lzy/桌面/OmniIsaacGymEnvs-main/omniisaacgymenvs/tasks/”, line 86, in post_reset
File “/home/lzy/桌面/OmniIsaacGymEnvs-main/omniisaacgymenvs/tasks/shared/”, line 178, in post_reset
File “/home/lzy/桌面/OmniIsaacGymEnvs-main/omniisaacgymenvs/tasks/shared/”, line 131, in reset_idx
dof_vel = torch_rand_float(-0.1, 0.1, (num_resets, self._robots.num_dof), device=self._device)
RuntimeError: nvrtc: error: invalid value for --gpu-architecture (-arch)

nvrtc compilation failed:

#define NAN __int_as_float(0x7fffffff)
#define POS_INFINITY __int_as_float(0x7f800000)
#define NEG_INFINITY __int_as_float(0xff800000)

device T maximum(T a, T b) {
return isnan(a) ? a : (a > b ? a : b);

device T minimum(T a, T b) {
return isnan(a) ? a : (a < b ? a : b);

extern “C” global
void fused_mul_add(double vlower_1, float* tv_, double vv__, float* aten_add) {
float v = ldg(tv + (long long)(threadIdx.x) + 512ll * (long long)(blockIdx.x));
aten_add[(long long)(threadIdx.x) + 512ll * (long long)(blockIdx.x)] = v * (float)(vv
_) + (float)(vlower_1);

Set the environment variable HYDRA_FULL_ERROR=1 for a complete stack trace.
/home/lzy/.local/share/ov/pkg/isaac_sim-2022.1.1/行 40: 5761 段错误 (核心已转储) $python_exe “$@” $args
There was an error running python

System and GPU info:
Ubuntu 22.04 LTS
NVIDIA-SMI 520.56.06
Driver Version: 520.56.06
CUDA Version: 11.8

Appreciate for your help QAQ

I have solved it! By install NVIDIA driver 525.53(beta). The problem comes from too low driver version.

I find that CUDA 11.8 need driver version >= 520.61.5, then I installed driver 525.53(beta) and it works. Using lower version CUDA like 11.7 may also work.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.