华为计算微信公众号
昇腾AI开发者公众号
华为计算微博
华为计算今日头条
4 台Atlas 800I A2 910B2,64G显存，使用mindie2.0.T3.1 ，以下关键参数
maxSeqLen：24576 maxInputTokenLen：16384 maxPrefillTokens = 16384 maxIterTimes:65535
可以正常使用。
想使用mindie2.1.RC1 进行推理。使用和 2.0T3.1 同样的配置，推理乱码无法使用，所以增加了以下几个dp,tp等配置，并调低maxSeqLen，maxInputTokenLen，maxPrefillTokens无法启动服务
"dp": 4, "tp": 8, "moe_tp": 1, "moe_ep": 32, "sp": 8
使用的配置如下：
export NPU_MEMORY_FRACTION=0.98
config.json
{ "Version": "1.0.0", "ServerConfig": { "ipAddress": "xxx", "managementIpAddress": "xxx", "port": 1025, "managementPort": 1026, "metricsPort": 1027, "allowAllZeroIpListening": false, "maxLinkNum": 1000, "httpsEnabled": false, "fullTextEnabled": false, "tlsCaPath": "security/ca/", "tlsCaFile": [ "ca.pem" ], "tlsCert": "security/certs/server.pem", "tlsPk": "security/keys/server.key.pem", "tlsPkPwd": "security/pass/key_pwd.txt", "tlsCrlPath": "security/certs/", "tlsCrlFiles": [ "server_crl.pem" ], "managementTlsCaFile": [ "management_ca.pem" ], "managementTlsCert": "security/certs/management/server.pem", "managementTlsPk": "security/keys/management/server.key.pem", "managementTlsPkPwd": "security/pass/management/key_pwd.txt", "managementTlsCrlPath": "security/management/certs/", "managementTlsCrlFiles": [ "server_crl.pem" ], "kmcKsfMaster": "tools/pmt/master/ksfa", "kmcKsfStandby": "tools/pmt/standby/ksfb", "inferMode": "standard", "interCommTLSEnabled": false, "interCommPort": 1121, "interCommTlsCaPath": "security/grpc/ca/", "interCommTlsCaFiles": [ "ca.pem" ], "interCommTlsCert": "security/grpc/certs/server.pem", "interCommPk": "security/grpc/keys/server.key.pem", "interCommPkPwd": "security/grpc/pass/key_pwd.txt", "interCommTlsCrlPath": "security/grpc/certs/", "interCommTlsCrlFiles": [ "server_crl.pem" ], "openAiSupport": "openai", "tokenTimeout": 600, "e2eTimeout": 600, "distDPServerEnabled": false }, "BackendConfig": { "backendName": "mindieservice_llm_engine", "modelInstanceNumber": 1, "npuDeviceIds": [ [ 0, 1, 2, 3, 4, 5, 6, 7 ] ], "tokenizerProcessNumber": 8, "multiNodesInferEnabled": true, "multiNodesInferPort": 1120, "interNodeTLSEnabled": false, "interNodeTlsCaPath": "security/grpc/ca/", "interNodeTlsCaFiles": [ "ca.pem" ], "interNodeTlsCert": "security/grpc/certs/server.pem", "interNodeTlsPk": "security/grpc/keys/server.key.pem", "interNodeTlsPkPwd": "security/grpc/pass/mindie_server_key_pwd.txt", "interNodeTlsCrlPath": "security/grpc/certs/", "interNodeTlsCrlFiles": [ "server_crl.pem" ], "interNodeKmcKsfMaster": "tools/pmt/master/ksfa", "interNodeKmcKsfStandby": "tools/pmt/standby/ksfb", "ModelDeployConfig": { "maxSeqLen": 16384, "maxInputTokenLen": 16384, "truncation": true, "ModelConfig": [ { "modelInstanceType": "Standard", "modelName": "deepseekr1", "modelWeightPath": "/models/DeepSeek-R1-0528-BF16", "worldSize": 8, "cpuMemSize": 5, "npuMemSize": -1, "backendType": "atb", "trustRemoteCode": true, "async_scheduler_wait_time": 120, "kv_trans_timeout": 10, "kv_link_timeout": 1080, "dp": 4, "tp": 8, "moe_tp": 1, "moe_ep": 32, "sp": 8 } ] }, "ScheduleConfig": { "templateType": "Standard", "templateName": "Standard_LLM", "cacheBlockSize": 128, "maxPrefillBatchSize": 50, "maxPrefillTokens": 16384, "prefillTimeMsPerReq": 150, "prefillPolicyType": 0, "decodeTimeMsPerReq": 50, "decodePolicyType": 0, "maxBatchSize": 200, "maxIterTimes": 16384, "maxPreemptCount": 0, "supportSelectBatch": false, "maxQueueDelayMicroseconds": 100000 } } }
报错信息如下：
/root/mindie/log/debug/mindie-batchscheduler_1180_20250829203355202.log:[2025-08-29 20:35:40.029+0800] [1180] [281472820769120] [batchscheduler] [ERROR] [model.py:61] : [Model] >>> Exception:NPU out of memory. Tried to allocate 17.14 GiB (NPU 0; 60.96 GiB total capacity; 44.50 GiB already allocated; 44.50 GiB current active; 15.59 GiB free; 44.54 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. /root/mindie/log/debug/mindie-batchscheduler_1180_20250829203355202.log:[2025-08-29 20:35:40.029+0800] [1180] [281472820769120] [batchscheduler] [ERROR] [model.py:64] : [MIE04E13030A] [Model] >>> return initialize error result: {'status': 'error', 'npuBlockNum': '0', 'cpuBlockNum': '0', 'memPoolId': '-1'} /root/mindie/log/debug/mindie-batchscheduler_1182_20250829203355213.log:[2025-08-29 20:35:40.029+0800] [1182] [281472678097248] [batchscheduler] [ERROR] [model.py:61] : [Model] >>> Exception:NPU out of memory. Tried to allocate 17.14 GiB (NPU 1; 60.96 GiB total capacity; 44.50 GiB already allocated; 44.50 GiB current active; 15.59 GiB free; 44.54 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. /root/mindie/log/debug/mindie-batchscheduler_1182_20250829203355213.log:[2025-08-29 20:35:40.029+0800] [1182] [281472678097248] [batchscheduler] [ERROR] [model.py:64] : [MIE04E13030A] [Model] >>> return initialize error result: {'status': 'error', 'npuBlockNum': '0', 'cpuBlockNum': '0', 'memPoolId': '-1'} /root/mindie/log/debug/mindie-batchscheduler_1184_20250829203355222.log:[2025-08-29 20:35:41.029+0800] [1184] [281472602534240] [batchscheduler] [ERROR] [model.py:61] : [Model] >>> Exception:NPU out of memory. Tried to allocate 17.14 GiB (NPU 2; 60.96 GiB total capacity; 44.50 GiB already allocated; 44.50 GiB current active; 15.59 GiB free; 44.54 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. /root/mindie/log/debug/mindie-batchscheduler_1184_20250829203355222.log:[2025-08-29 20:35:41.029+0800] [1184] [281472602534240] [batchscheduler] [ERROR] [model.py:64] : [MIE04E13030A] [Model] >>> return initialize error result: {'status': 'error', 'npuBlockNum': '0', 'cpuBlockNum': '0', 'memPoolId': '-1'} /root/mindie/log/debug/mindie-batchscheduler_1194_20250829203355232.log:[2025-08-29 20:35:41.029+0800] [1194] [281472610988384] [batchscheduler] [ERROR] [model.py:61] : [Model] >>> Exception:NPU out of memory. Tried to allocate 17.14 GiB (NPU 3; 60.96 GiB total capacity; 44.50 GiB already allocated; 44.50 GiB current active; 15.59 GiB free; 44.54 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. /root/mindie/log/debug/mindie-batchscheduler_1194_20250829203355232.log:[2025-08-29 20:35:41.029+0800] [1194] [281472610988384] [batchscheduler] [ERROR] [model.py:64] : [MIE04E13030A] [Model] >>> return initialize error result: {'status': 'error', 'npuBlockNum': '0', 'cpuBlockNum': '0', 'memPoolId': '-1'} /root/mindie/log/debug/mindie-batchscheduler_1204_20250829203355243.log:[2025-08-29 20:35:41.029+0800] [1204] [281472652996960] [batchscheduler] [ERROR] [model.py:61] : [Model] >>> Exception:NPU out of memory. Tried to allocate 17.14 GiB (NPU 4; 60.96 GiB total capacity; 44.50 GiB already allocated; 44.50 GiB current active; 15.59 GiB free; 44.54 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. /root/mindie/log/debug/mindie-batchscheduler_1204_20250829203355243.log:[2025-08-29 20:35:41.029+0800] [1204] [281472652996960] [batchscheduler] [ERROR] [model.py:64] : [MIE04E13030A] [Model] >>> return initialize error result: {'status': 'error', 'npuBlockNum': '0', 'cpuBlockNum': '0', 'memPoolId': '-1'} /root/mindie/log/debug/mindie-batchscheduler_1214_20250829203355254.log:[2025-08-29 20:35:41.029+0800] [1214] [281473458303328] [batchscheduler] [ERROR] [model.py:61] : [Model] >>> Exception:NPU out of memory. Tried to allocate 17.14 GiB (NPU 5; 60.96 GiB total capacity; 44.50 GiB already allocated; 44.50 GiB current active; 15.59 GiB free; 44.54 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. /root/mindie/log/debug/mindie-batchscheduler_1214_20250829203355254.log:[2025-08-29 20:35:41.029+0800] [1214] [281473458303328] [batchscheduler] [ERROR] [model.py:64] : [MIE04E13030A] [Model] >>> return initialize error result: {'status': 'error', 'npuBlockNum': '0', 'cpuBlockNum': '0', 'memPoolId': '-1'} /root/mindie/log/debug/mindie-batchscheduler_1224_20250829203355265.log:[2025-08-29 20:35:41.029+0800] [1224] [281473416294752] [batchscheduler] [ERROR] [model.py:61] : [Model] >>> Exception:NPU out of memory. Tried to allocate 17.14 GiB (NPU 6; 60.96 GiB total capacity; 44.50 GiB already allocated; 44.50 GiB current active; 15.59 GiB free; 44.54 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. /root/mindie/log/debug/mindie-batchscheduler_1224_20250829203355265.log:[2025-08-29 20:35:41.029+0800] [1224] [281473416294752] [batchscheduler] [ERROR] [model.py:64] : [MIE04E13030A] [Model] >>> return initialize error result: {'status': 'error', 'npuBlockNum': '0', 'cpuBlockNum': '0', 'memPoolId': '-1'} /root/mindie/log/debug/mindie-batchscheduler_1235_20250829203355281.log:[2025-08-29 20:35:41.029+0800] [1235] [281472703197536] [batchscheduler] [ERROR] [model.py:61] : [Model] >>> Exception:NPU out of memory. Tried to allocate 17.14 GiB (NPU 7; 60.96 GiB total capacity; 44.50 GiB already allocated; 44.50 GiB current active; 15.59 GiB free; 44.54 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. /root/mindie/log/debug/mindie-batchscheduler_1235_20250829203355281.log:[2025-08-29 20:35:41.029+0800] [1235] [281472703197536] [batchscheduler] [ERROR] [model.py:64] : [MIE04E13030A] [Model] >>> return initialize error result: {'status': 'error', 'npuBlockNum': '0', 'cpuBlockNum': '0', 'memPoolId': '-1'} /root/mindie/log/debug/mindie-llm_1180_20250829203401175.log:[2025-08-29 20:35:40,691] [1180] [281470003048800] [llm] [ERROR] [logging.py-346] : Warmup failed, because of model inference out of memory when `npuMemSize` set to -1. please try to decrease `npuMemSize` /root/mindie/log/debug/mindie-server_1133_202508292035099.log:[2025-08-29 20:35:41.100+08:00] [1133] [1137] [server] [ERROR] [llm_daemon.cpp:90] : [MIE04E010109] [daemon] Process 1182 was terminated by signal 9 (Killed) /root/mindie/log/debug/mindie-server_1133_202508292035099.log:[2025-08-29 20:35:41.153+08:00] [1133] [1133] [server] [ERROR] [llm_daemon.cpp:90] : [MIE04E010109] [daemon] Process 1180 was terminated by signal 9 (Killed)
我要发帖子
点赞
收藏
回复
分享
举报
已解决 已解决
4 台Atlas 800I A2 910B2,64G显存，使用mindie2.0.T3.1 ，以下关键参数
可以正常使用。
想使用mindie2.1.RC1 进行推理。使用和 2.0T3.1 同样的配置，推理乱码无法使用，所以增加了以下几个dp,tp等配置，并调低maxSeqLen，maxInputTokenLen，maxPrefillTokens无法启动服务
使用的配置如下：
config.json
{ "Version": "1.0.0", "ServerConfig": { "ipAddress": "xxx", "managementIpAddress": "xxx", "port": 1025, "managementPort": 1026, "metricsPort": 1027, "allowAllZeroIpListening": false, "maxLinkNum": 1000, "httpsEnabled": false, "fullTextEnabled": false, "tlsCaPath": "security/ca/", "tlsCaFile": [ "ca.pem" ], "tlsCert": "security/certs/server.pem", "tlsPk": "security/keys/server.key.pem", "tlsPkPwd": "security/pass/key_pwd.txt", "tlsCrlPath": "security/certs/", "tlsCrlFiles": [ "server_crl.pem" ], "managementTlsCaFile": [ "management_ca.pem" ], "managementTlsCert": "security/certs/management/server.pem", "managementTlsPk": "security/keys/management/server.key.pem", "managementTlsPkPwd": "security/pass/management/key_pwd.txt", "managementTlsCrlPath": "security/management/certs/", "managementTlsCrlFiles": [ "server_crl.pem" ], "kmcKsfMaster": "tools/pmt/master/ksfa", "kmcKsfStandby": "tools/pmt/standby/ksfb", "inferMode": "standard", "interCommTLSEnabled": false, "interCommPort": 1121, "interCommTlsCaPath": "security/grpc/ca/", "interCommTlsCaFiles": [ "ca.pem" ], "interCommTlsCert": "security/grpc/certs/server.pem", "interCommPk": "security/grpc/keys/server.key.pem", "interCommPkPwd": "security/grpc/pass/key_pwd.txt", "interCommTlsCrlPath": "security/grpc/certs/", "interCommTlsCrlFiles": [ "server_crl.pem" ], "openAiSupport": "openai", "tokenTimeout": 600, "e2eTimeout": 600, "distDPServerEnabled": false }, "BackendConfig": { "backendName": "mindieservice_llm_engine", "modelInstanceNumber": 1, "npuDeviceIds": [ [ 0, 1, 2, 3, 4, 5, 6, 7 ] ], "tokenizerProcessNumber": 8, "multiNodesInferEnabled": true, "multiNodesInferPort": 1120, "interNodeTLSEnabled": false, "interNodeTlsCaPath": "security/grpc/ca/", "interNodeTlsCaFiles": [ "ca.pem" ], "interNodeTlsCert": "security/grpc/certs/server.pem", "interNodeTlsPk": "security/grpc/keys/server.key.pem", "interNodeTlsPkPwd": "security/grpc/pass/mindie_server_key_pwd.txt", "interNodeTlsCrlPath": "security/grpc/certs/", "interNodeTlsCrlFiles": [ "server_crl.pem" ], "interNodeKmcKsfMaster": "tools/pmt/master/ksfa", "interNodeKmcKsfStandby": "tools/pmt/standby/ksfb", "ModelDeployConfig": { "maxSeqLen": 16384, "maxInputTokenLen": 16384, "truncation": true, "ModelConfig": [ { "modelInstanceType": "Standard", "modelName": "deepseekr1", "modelWeightPath": "/models/DeepSeek-R1-0528-BF16", "worldSize": 8, "cpuMemSize": 5, "npuMemSize": -1, "backendType": "atb", "trustRemoteCode": true, "async_scheduler_wait_time": 120, "kv_trans_timeout": 10, "kv_link_timeout": 1080, "dp": 4, "tp": 8, "moe_tp": 1, "moe_ep": 32, "sp": 8 } ] }, "ScheduleConfig": { "templateType": "Standard", "templateName": "Standard_LLM", "cacheBlockSize": 128, "maxPrefillBatchSize": 50, "maxPrefillTokens": 16384, "prefillTimeMsPerReq": 150, "prefillPolicyType": 0, "decodeTimeMsPerReq": 50, "decodePolicyType": 0, "maxBatchSize": 200, "maxIterTimes": 16384, "maxPreemptCount": 0, "supportSelectBatch": false, "maxQueueDelayMicroseconds": 100000 } } }报错信息如下：
/root/mindie/log/debug/mindie-batchscheduler_1180_20250829203355202.log:[2025-08-29 20:35:40.029+0800] [1180] [281472820769120] [batchscheduler] [ERROR] [model.py:61] : [Model] >>> Exception:NPU out of memory. Tried to allocate 17.14 GiB (NPU 0; 60.96 GiB total capacity; 44.50 GiB already allocated; 44.50 GiB current active; 15.59 GiB free; 44.54 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. /root/mindie/log/debug/mindie-batchscheduler_1180_20250829203355202.log:[2025-08-29 20:35:40.029+0800] [1180] [281472820769120] [batchscheduler] [ERROR] [model.py:64] : [MIE04E13030A] [Model] >>> return initialize error result: {'status': 'error', 'npuBlockNum': '0', 'cpuBlockNum': '0', 'memPoolId': '-1'} /root/mindie/log/debug/mindie-batchscheduler_1182_20250829203355213.log:[2025-08-29 20:35:40.029+0800] [1182] [281472678097248] [batchscheduler] [ERROR] [model.py:61] : [Model] >>> Exception:NPU out of memory. Tried to allocate 17.14 GiB (NPU 1; 60.96 GiB total capacity; 44.50 GiB already allocated; 44.50 GiB current active; 15.59 GiB free; 44.54 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. /root/mindie/log/debug/mindie-batchscheduler_1182_20250829203355213.log:[2025-08-29 20:35:40.029+0800] [1182] [281472678097248] [batchscheduler] [ERROR] [model.py:64] : [MIE04E13030A] [Model] >>> return initialize error result: {'status': 'error', 'npuBlockNum': '0', 'cpuBlockNum': '0', 'memPoolId': '-1'} /root/mindie/log/debug/mindie-batchscheduler_1184_20250829203355222.log:[2025-08-29 20:35:41.029+0800] [1184] [281472602534240] [batchscheduler] [ERROR] [model.py:61] : [Model] >>> Exception:NPU out of memory. Tried to allocate 17.14 GiB (NPU 2; 60.96 GiB total capacity; 44.50 GiB already allocated; 44.50 GiB current active; 15.59 GiB free; 44.54 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. /root/mindie/log/debug/mindie-batchscheduler_1184_20250829203355222.log:[2025-08-29 20:35:41.029+0800] [1184] [281472602534240] [batchscheduler] [ERROR] [model.py:64] : [MIE04E13030A] [Model] >>> return initialize error result: {'status': 'error', 'npuBlockNum': '0', 'cpuBlockNum': '0', 'memPoolId': '-1'} /root/mindie/log/debug/mindie-batchscheduler_1194_20250829203355232.log:[2025-08-29 20:35:41.029+0800] [1194] [281472610988384] [batchscheduler] [ERROR] [model.py:61] : [Model] >>> Exception:NPU out of memory. Tried to allocate 17.14 GiB (NPU 3; 60.96 GiB total capacity; 44.50 GiB already allocated; 44.50 GiB current active; 15.59 GiB free; 44.54 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. /root/mindie/log/debug/mindie-batchscheduler_1194_20250829203355232.log:[2025-08-29 20:35:41.029+0800] [1194] [281472610988384] [batchscheduler] [ERROR] [model.py:64] : [MIE04E13030A] [Model] >>> return initialize error result: {'status': 'error', 'npuBlockNum': '0', 'cpuBlockNum': '0', 'memPoolId': '-1'} /root/mindie/log/debug/mindie-batchscheduler_1204_20250829203355243.log:[2025-08-29 20:35:41.029+0800] [1204] [281472652996960] [batchscheduler] [ERROR] [model.py:61] : [Model] >>> Exception:NPU out of memory. Tried to allocate 17.14 GiB (NPU 4; 60.96 GiB total capacity; 44.50 GiB already allocated; 44.50 GiB current active; 15.59 GiB free; 44.54 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. /root/mindie/log/debug/mindie-batchscheduler_1204_20250829203355243.log:[2025-08-29 20:35:41.029+0800] [1204] [281472652996960] [batchscheduler] [ERROR] [model.py:64] : [MIE04E13030A] [Model] >>> return initialize error result: {'status': 'error', 'npuBlockNum': '0', 'cpuBlockNum': '0', 'memPoolId': '-1'} /root/mindie/log/debug/mindie-batchscheduler_1214_20250829203355254.log:[2025-08-29 20:35:41.029+0800] [1214] [281473458303328] [batchscheduler] [ERROR] [model.py:61] : [Model] >>> Exception:NPU out of memory. Tried to allocate 17.14 GiB (NPU 5; 60.96 GiB total capacity; 44.50 GiB already allocated; 44.50 GiB current active; 15.59 GiB free; 44.54 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. /root/mindie/log/debug/mindie-batchscheduler_1214_20250829203355254.log:[2025-08-29 20:35:41.029+0800] [1214] [281473458303328] [batchscheduler] [ERROR] [model.py:64] : [MIE04E13030A] [Model] >>> return initialize error result: {'status': 'error', 'npuBlockNum': '0', 'cpuBlockNum': '0', 'memPoolId': '-1'} /root/mindie/log/debug/mindie-batchscheduler_1224_20250829203355265.log:[2025-08-29 20:35:41.029+0800] [1224] [281473416294752] [batchscheduler] [ERROR] [model.py:61] : [Model] >>> Exception:NPU out of memory. Tried to allocate 17.14 GiB (NPU 6; 60.96 GiB total capacity; 44.50 GiB already allocated; 44.50 GiB current active; 15.59 GiB free; 44.54 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. /root/mindie/log/debug/mindie-batchscheduler_1224_20250829203355265.log:[2025-08-29 20:35:41.029+0800] [1224] [281473416294752] [batchscheduler] [ERROR] [model.py:64] : [MIE04E13030A] [Model] >>> return initialize error result: {'status': 'error', 'npuBlockNum': '0', 'cpuBlockNum': '0', 'memPoolId': '-1'} /root/mindie/log/debug/mindie-batchscheduler_1235_20250829203355281.log:[2025-08-29 20:35:41.029+0800] [1235] [281472703197536] [batchscheduler] [ERROR] [model.py:61] : [Model] >>> Exception:NPU out of memory. Tried to allocate 17.14 GiB (NPU 7; 60.96 GiB total capacity; 44.50 GiB already allocated; 44.50 GiB current active; 15.59 GiB free; 44.54 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. /root/mindie/log/debug/mindie-batchscheduler_1235_20250829203355281.log:[2025-08-29 20:35:41.029+0800] [1235] [281472703197536] [batchscheduler] [ERROR] [model.py:64] : [MIE04E13030A] [Model] >>> return initialize error result: {'status': 'error', 'npuBlockNum': '0', 'cpuBlockNum': '0', 'memPoolId': '-1'} /root/mindie/log/debug/mindie-llm_1180_20250829203401175.log:[2025-08-29 20:35:40,691] [1180] [281470003048800] [llm] [ERROR] [logging.py-346] : Warmup failed, because of model inference out of memory when `npuMemSize` set to -1. please try to decrease `npuMemSize` /root/mindie/log/debug/mindie-server_1133_202508292035099.log:[2025-08-29 20:35:41.100+08:00] [1133] [1137] [server] [ERROR] [llm_daemon.cpp:90] : [MIE04E010109] [daemon] Process 1182 was terminated by signal 9 (Killed) /root/mindie/log/debug/mindie-server_1133_202508292035099.log:[2025-08-29 20:35:41.153+08:00] [1133] [1133] [server] [ERROR] [llm_daemon.cpp:90] : [MIE04E010109] [daemon] Process 1180 was terminated by signal 9 (Killed)