{"record":{"id":"bed4728374d61fe2","repo":"vllm-project/vllm","slug":"cannot-use-renderer-num-workers-1-with-the-mul","errorCode":null,"errorMessage":"Cannot use --renderer-num-workers > 1 with the multimodal processor cache enabled for pooling models. Pooling preprocessing runs on the renderer workers, and the cache is not thread-safe. Please set --renderer-num-workers 1 (the default), or disable the cache with --mm-processor-cache-gb 0.","messagePattern":"Cannot use --renderer-num-workers > 1 with the multimodal processor cache enabled for pooling models\\. Pooling preprocessing runs on the renderer workers, and the cache is not thread-safe\\. Please set --renderer-num-workers 1 \\(the default\\), or disable the cache with --mm-processor-cache-gb 0\\.","errorType":"validation","errorClass":"ValueError","httpStatus":null,"severity":"error","filePath":"vllm/config/model.py","lineNumber":822,"sourceCode":"            pruning_spec = self.multimodal_config.get_video_pruning_spec()\n            supported_pruning = self._model_info.supported_video_pruning_methods\n            if (\n                pruning_spec is not None\n                and supported_pruning\n                and pruning_spec[0] not in supported_pruning\n            ):\n                raise ValueError(\n                    f\"Video pruning method '{pruning_spec[0]}' is not \"\n                    f\"supported by {self._model_info.architecture} \"\n                    f\"(supported methods: {supported_pruning}).\"\n                )\n\n            if (\n                self.renderer_num_workers > 1\n                and self.multimodal_config.mm_processor_cache_gb > 0\n                and self.runner_type == \"pooling\"\n            ):\n                raise ValueError(\n                    \"Cannot use --renderer-num-workers > 1 with the \"\n                    \"multimodal processor cache enabled for pooling models. \"\n                    \"Pooling preprocessing runs on the renderer workers, and \"\n                    \"the cache is not thread-safe. Please set \"\n                    \"--renderer-num-workers 1 (the default), or \"\n                    \"disable the cache with --mm-processor-cache-gb 0.\"\n                )\n\n            # Rebuild after multimodal_config exists so text-only mm_prefix\n            # clearing is applied (and cached for later with_hf_config calls).\n            self.model_arch_config = self.get_model_arch_config()\n\n        if self.disable_sliding_window:\n            # Set after get_and_verify_max_len to ensure that max_model_len\n            # can be correctly capped to sliding window size\n            self.hf_text_config.sliding_window = None\n\n        # Avoid running try_verify_and_update_config multiple times","sourceCodeStart":804,"sourceCodeEnd":840,"githubUrl":"https://github.com/vllm-project/vllm/blob/c794754062d49a8fdb63ab3c5215b488b865030c/vllm/config/model.py#L804-L840","documentation":"Error \"Cannot use --renderer-num-workers > 1 with the multimodal processor cache enabled for pooling models. Pooling preprocessing runs on the renderer workers, and the cache is not thread-safe. Please set --renderer-num-workers 1 (the default), or disable the cache with --mm-processor-cache-gb 0.\" thrown in vllm-project/vllm.","triggerScenarios":"Raised at vllm/config/model.py:822 when validation fails: Cannot use --renderer-num-workers > 1 with the multimodal processor cache enabled for pooling models. 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/model.py:822 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":["Set --renderer-num-workers 1 (the default) for pooling models with the multimodal processor cache enabled.","Or disable the cache with --mm-processor-cache-gb 0."],"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"}