{
  "id": "cuda/device-incompatible-cubin",
  "signature": "CUDA error: no kernel image is available for execution on the device (cudaErrorNoKernelImageAvailable)",
  "signature_zh": "CUDA错误：没有可用于在设备上执行的内核映像 (cudaErrorNoKernelImageAvailable)",
  "regex": "CUDA error: no kernel image is available for execution on the device",
  "domain": "cuda",
  "category": "build_error",
  "subcategory": null,
  "root_cause": "The compiled CUDA binary (cubin) was generated for a different GPU architecture than the one currently in use, e.g., sm_80 vs sm_86.",
  "root_cause_type": "generic",
  "root_cause_zh": "编译的CUDA二进制文件（cubin）是为与当前使用的GPU架构不同的架构生成的，例如sm_80与sm_86。",
  "versions": [
    {
      "version": "CUDA 12.0",
      "introduced": null,
      "deprecated": null,
      "removed": null,
      "behavior_change": null,
      "status": "active"
    },
    {
      "version": "PyTorch 2.0.0",
      "introduced": null,
      "deprecated": null,
      "removed": null,
      "behavior_change": null,
      "status": "active"
    },
    {
      "version": "NVIDIA A100 (sm_80)",
      "introduced": null,
      "deprecated": null,
      "removed": null,
      "behavior_change": null,
      "status": "active"
    }
  ],
  "os_specific": {},
  "dead_ends": [
    {
      "action": "",
      "why_fails": "The issue is architecture-specific, not CUDA version.",
      "fail_rate": 0.8,
      "condition": "",
      "sources": []
    },
    {
      "action": "",
      "why_fails": "Caching only affects runtime memory, not compiled code.",
      "fail_rate": 0.95,
      "condition": "",
      "sources": []
    }
  ],
  "workarounds": [
    {
      "action": "Recompile the kernel with the correct architecture. For PyTorch, set environment variable `TORCH_CUDA_ARCH_LIST='8.6'` for Ampere GPUs. For custom CUDA code, compile with `-arch=sm_86`.",
      "success_rate": 0.9,
      "how": "Recompile the kernel with the correct architecture. For PyTorch, set environment variable `TORCH_CUDA_ARCH_LIST='8.6'` for Ampere GPUs. For custom CUDA code, compile with `-arch=sm_86`.",
      "condition": "",
      "sources": []
    }
  ],
  "workarounds_zh": [
    "Recompile the kernel with the correct architecture. For PyTorch, set environment variable `TORCH_CUDA_ARCH_LIST='8.6'` for Ampere GPUs. For custom CUDA code, compile with `-arch=sm_86`."
  ],
  "transition_graph": {
    "leads_to": [],
    "preceded_by": [],
    "frequently_confused_with": []
  },
  "official_doc_url": "https://docs.nvidia.com/cuda/cuda-compiler-driver-nvcc/index.html#gpu-compilation",
  "official_doc_section": null,
  "error_code": "cudaErrorNoKernelImageAvailable",
  "verification_tier": "ai_generated",
  "confidence": 0.88,
  "fix_success_rate": 0.9,
  "resolvable": "true",
  "first_seen": "2024-01-10",
  "last_confirmed": "2024-06-01",
  "last_updated": "2024-06-01",
  "evidence_count": 1,
  "tags": [],
  "locale": "en",
  "aliases": []
}