{"record":{"id":"9411f959a68bd3b1","repo":"vllm-project/vllm","slug":"custom-ops-must-contain-exactly-one-base-mode-al","errorCode":null,"errorMessage":"custom_ops must contain exactly one base mode: 'all' or 'none'","messagePattern":"custom_ops must contain exactly one base mode: 'all' or 'none'","errorType":"validation","errorClass":"ValueError","httpStatus":null,"severity":"error","filePath":"vllm/config/compilation.py","lineNumber":1360,"sourceCode":"                    \"doesn't exist (or wasn't imported/registered)\"\n                    if op_name not in op_registry\n                    else \"not present in model\"\n                )\n\n                enable_str = \"enabling\" if op[0] == \"+\" else \"disabling\"\n                logger.warning_once(\n                    \"Op '%s' %s, %s with '%s' has no effect\",\n                    op_name,\n                    missing_str,\n                    enable_str,\n                    op,\n                )\n\n    def is_custom_op_enabled(self, op: str) -> bool:\n        count_all = self.custom_ops.count(\"all\")\n        count_none = self.custom_ops.count(\"none\")\n        if count_all + count_none != 1:\n            raise ValueError(\n                \"custom_ops must contain exactly one base mode: 'all' or 'none'\"\n            )\n\n        if count_all:\n            return f\"-{op}\" not in self.custom_ops\n\n        return f\"+{op}\" in self.custom_ops\n\n    def resolve_cudagraph_mode_and_sizes(\n        self,\n        min_cg_support: \"AttentionCGSupport\",\n        min_cg_attn_backend: str | None,\n        uniform_decode_query_len: int = 1,\n        use_v2_model_runner: bool = False,\n        tensor_parallel_size: int = 1,\n        kv_cache_config: \"KVCacheConfig | None\" = None,\n        max_num_reqs: int | None = None,\n        is_profiling: bool = False,","sourceCodeStart":1342,"sourceCodeEnd":1378,"githubUrl":"https://github.com/vllm-project/vllm/blob/c794754062d49a8fdb63ab3c5215b488b865030c/vllm/config/compilation.py#L1342-L1378","documentation":"Error \"custom_ops must contain exactly one base mode: 'all' or 'none'\" thrown in vllm-project/vllm.","triggerScenarios":"Raised at vllm/config/compilation.py:1360 when validation fails: custom_ops must contain exactly one base mode: 'all' or 'none'. Typically triggered by an incompatible or incomplete vLLM configuration, an unsupported platform/backend combination, or a runtime resource/dependency that is missing.","commonSituations":"Commonly encountered at vllm/config/compilation.py:1360 during vLLM startup/config validation or runtime setup when: (1) conflicting CLI flags or config fields are combined, (2) the current platform (CUDA/ROCm/CPU/XPU) or installed optional packages do not support the requested feature, or (3) a required value is absent or out of range. Resolve by correcting the configuration as described in the message, or by selecting a supported alternative.","solutions":["Include exactly one base mode in custom_ops: 'all' or 'none'.","Example: --compilation-config '{\"custom_ops\":[\"all\",\"-silu_and_mul\"]}'."],"exampleFix":null,"handlingStrategy":null,"validationCode":null,"typeGuard":null,"tryCatchPattern":null,"preventionTips":[],"tags":[],"backgroundTag":null,"analyzedSha":"c794754062d49a8fdb63ab3c5215b488b865030c","analyzedAt":"2026-08-14T21:17:39.825Z","schemaVersion":2},"datasetVersion":"2026-08-15T17:31:12.345Z"}