CUDA_ERROR_INSUFFICIENT_DRIVER warning CUDA driver version mismatch
Installed CUDA toolkit version is newer than what the installed driver supports.
Error directory
Common failures during local inference, grouped by the module that throws them.
CUDA
CUDA_ERROR_OUT_OF_MEMORY critical CUDA out of memory
The model, context, or batch size requested more VRAM than the GPU has available.
Ollama
OLLAMA_MODEL_NOT_FOUND info Ollama model not found locally
The model tag hasn't been pulled yet, or the tag name doesn't match the registry exactly.
vLLM
VLLM_KV_CACHE_OOM warning vLLM KV cache allocation failure
gpu_memory_utilization is set too high, leaving no room for the KV cache.
VLLM_TENSOR_PARALLEL_CRASH critical vLLM tensor-parallel crash on startup
Tensor-parallel size doesn't evenly divide the model's attention heads.