Hugging Face's logo
cmeow23
/
Agents
Runtime error

runtime error

Exit code: 1. Reason: /site-packages/zonos2/engine/engine.py", line 19, in <module> from zonos2.models import create_model, load_checkpoint_weight File "/usr/local/lib/python3.10/site-packages/zonos2/models/__init__.py", line 5, in <module> from .zonos2 import Zonos2ForCausalLM File "/usr/local/lib/python3.10/site-packages/zonos2/models/zonos2.py", line 23, in <module> from zonos2.layers.moe.fused_moe.layer import FusedMoE File "/usr/local/lib/python3.10/site-packages/zonos2/layers/moe/fused_moe/layer.py", line 7, in <module> from zonos2.layers.moe.fused_moe.fused_moe_impl import fused_experts, fused_moe File "/usr/local/lib/python3.10/site-packages/zonos2/layers/moe/fused_moe/fused_moe_impl.py", line 7, in <module> from sgl_kernel import gelu_and_mul, silu_and_mul File "/usr/local/lib/python3.10/site-packages/sgl_kernel/__init__.py", line 5, in <module> common_ops = _load_architecture_specific_ops() File "/usr/local/lib/python3.10/site-packages/sgl_kernel/load_utils.py", line 188, in _load_architecture_specific_ops raise ImportError(error_msg) ImportError: [sgl_kernel] CRITICAL: Could not load any common_ops library! Attempted locations: 1. Architecture-specific pattern: /usr/local/lib/python3.10/site-packages/sgl_kernel/sm100/common_ops.* - found files: ['/usr/local/lib/python3.10/site-packages/sgl_kernel/sm100/common_ops.abi3.so'] 2. Fallback pattern: /usr/local/lib/python3.10/site-packages/sgl_kernel/common_ops.* - found files: [] 3. Standard Python import: common_ops - failed GPU Info: - Compute capability: None - Expected variant: CPU/No GPU detected (using precise math) Please ensure sgl_kernel is properly installed with: pip install --upgrade sgl_kernel Error details from previous import attempts: - ImportError: libcuda.so.1: cannot open shared object file: No such file or directory - ModuleNotFoundError: No module named 'common_ops' [W616 12:25:25.679397521 AllocatorConfig.cpp:28] Warning: PYTORCH_CUDA_ALLOC_CONF is deprecated, use PYTORCH_ALLOC_CONF instead (function operator())

Container logs:

Fetching error logs...