1. 接口描述
接口请求域名： dlc.tencentcloudapi.com 。
就地更新训练作业配置
默认接口请求频率限制：20次/秒。
推荐使用 API Explorer
点击调试
API Explorer 提供了在线调用、签名验证、SDK 代码生成和快速检索接口等能力。您可查看每次调用的请求内容和返回结果以及自动生成 SDK 调用示例。
2. 输入参数
以下请求参数列表仅列出了接口请求参数和部分公共参数，完整公共参数列表见 公共请求参数。
| 参数名称 | 必选 | 类型 | 描述 |
|---|---|---|---|
| Action | 是 | String | 公共参数，本接口取值：ModifyTrainingJobSpec。 |
| Version | 是 | String | 公共参数，本接口取值：2021-01-25。 |
| Region | 是 | String | 公共参数，详见产品支持的 地域列表，本接口仅支持其中的: ap-beijing, ap-chongqing, ap-guangzhou, ap-shanghai, ap-singapore 。 |
| SpecId | 否 | String | 配置 ID 示例值：raytrain-spec-tiogqf-yb6e |
| SpecName | 否 | String | 配置名称（不传则不更新） 示例值：xxd |
| Description | 否 | String | 配置描述（不传则不更新） 示例值：sadfafda |
| Entrypoint | 否 | String | 启动命令（不传则不更新） 示例值：bash -c 'cd sft_demo && python -m sft_demo.train_local --storage-path /shared' |
| Image | 否 | String | 镜像地址（不传则不更新） 示例值：ccr.ccs.tencentyun.com/emr-image/tcray:2.55.1-py311-cpu-torch-lab |
| ImagePullType | 否 | String | 镜像拉取类型（BuiltIn / Custom / CustomCcr，不传则不更新） 示例值：BuiltIn |
| ImagePullPolicy | 否 | String | 镜像拉取策略（Always / IfNotPresent / Never，不传则不更新） 示例值：IfNotPresent |
| CodePackageUrl | 否 | String | 代码包 COS URL（不传则不更新） 示例值：https://common-job-packages-251233710.cos.ap-guangzhou.myqcloud.com/job-packages/260200065/20260630194144/44a9489e-e60d-4d73-a048-8a5fe39b2d32.zip |
| RuntimeEnv | 否 | String | 运行时环境配置 JSON（不传则不更新） 示例值：{ "env_vars": { "LEARNING_RATE": "1e-4", "BATCH_SIZE": "32" }, "pip": [] } |
| ResourceConfigId | 否 | String | 资源配置模板 ID（可选） 示例值：0453d291-7b46-4b04-a989-f659d419b8a5 |
| ResourceConfig | 否 | String | 资源配置 JSON（不传则不更新） 示例值：{"Head":{"Name":"default-head","PodCpu":2,"PodMem":8,"GpuType":"","GpuNum":0,"Envs":[],"PodNum":1,"HighAvailability":false},"Worker":[{"Name":"default-worker","PodCpu":2,"PodMem":8,"GpuType":"","GpuNum":0,"Envs":[],"MinPodNum":1,"MaxPodNum":1}]} |
| ResourcePartitionId | 否 | String | 资源分区 ID（不传则不更新） 示例值：dlc-p-wdtiljwu |
| Queue | 否 | String | 队列名称（不传则不更新） 示例值：default |
| Catalog | 否 | String | 存储卷挂载配置 JSON（不传则不更新） 示例值：{"COSVolumes":[{"Bucket":"xx","MountPath":"/x","VolumeSubPath":"/xx","Source":"dataset"}],"CFSVolumes":[{"FileSystemId":"xx","MountPath":"/x","VolumeSubPath":"/xx","AutoCreateSubDir":true,"Source":"checkpoint"}]} |
| Priority | 否 | Integer | 作业优先级 1-9（不传则不更新） 示例值：5 |
| AdvancedOptions | 否 | String | 高级参数 JSON（不传则不更新） 示例值：{"spec.suspend":"false"} |
| MlFlowConfig | 否 | MlFlowConfig | MlFlow 实验追踪配置（不传则不更新） |
| Tags.N | 否 | Array of Tag | 标签列表（TagKey-TagValue），null 不修改，空数组清空，非空全量替换 |
| BaseModelUid | 否 | String | 基础模型ID（用于模型挂载） 示例值：m-test1-6a1e7ed1-33de |
| OutputModelName | 否 | String | 输出模型名（用于产出模型自动注册） 示例值：Qwen-0.5B |
| Mode | 否 | String | 训练模式：sft / dpo / cpt / grpo（仅 POST_TRAINING 有值） 示例值：sft |
| Datasets.N | 否 | Array of DatasetMount | 数据集挂载列表（整体替换，不传则不更新） |
| Checkpoint | 否 | CheckpointConfig | Checkpoint 产出配置（整体替换，不传则不更新） |
| ResumeTraining | 否 | Boolean | 是否启用断点续训（创建时的意图声明；实际续训由实例级「断点续训」按钮触发，不传则不更新） 示例值：false |
| TuningParams | 否 | TrainingTuningParams | 调优参数（整体替换，未填字段回模板默认值；不传则不更新；仅 POST_TRAINING） |
3. 输出参数
| 参数名称 | 类型 | 描述 |
|---|---|---|
| Spec | TrainingJobSpec | 训练作业配置详情 |
| RequestId | String | 唯一请求 ID，由服务端生成，每次请求都会返回（若请求因其他原因未能抵达服务端，则该次请求不会获得 RequestId）。定位问题时需要提供该次请求的 RequestId。 |
4. 示例
示例1 更新训练作业配置
输入示例
POST / HTTP/1.1
Host: dlc.tencentcloudapi.com
Content-Type: application/json
X-TC-Action: ModifyTrainingJobSpec
<公共请求参数>
{
"SpecId": "raytrain-spec-tiogqf-yb6e",
"SpecName": "xxd",
"Description": "sadfafda"
}输出示例
{
"Response": {
"Spec": {
"BaseModelName": "Qwen3.5-4B",
"BaseModelUid": "m-qwen3-5-4b-6a3218e7-7447",
"Catalog": "{\"CFSVolumes\":null,\"CFSTurboVolumes\":null,\"COSVolumes\":[{\"Region\":\"ap-guangzhou\",\"Bucket\":\"common-job-packages-251233710\",\"VolumeSubPath\":\"/builtin-models/Qwen3.5-4B/v1/\",\"SubPathMode\":null,\"SubPath\":null,\"MountPath\":\"/builtin-models/Qwen3.5-4B/v1\",\"PersistVolumeName\":null,\"VolumeMountMode\":\"ReadOnly\",\"Source\":null},{\"Region\":null,\"Bucket\":\"aidanyxu-cos-260200065\",\"VolumeSubPath\":\"\",\"SubPathMode\":null,\"SubPath\":null,\"MountPath\":\"/\",\"PersistVolumeName\":null,\"VolumeMountMode\":\"ReadWrite\",\"Source\":\"checkpoint\"},{\"Region\":null,\"Bucket\":\"aidanyxu-cos-260200065\",\"VolumeSubPath\":\"\",\"SubPathMode\":\"subPath\",\"SubPath\":\"\",\"MountPath\":\"\",\"PersistVolumeName\":null,\"VolumeMountMode\":null,\"Source\":\"dataset\"}],\"GooseFSVolumes\":null}",
"Checkpoint": {
"Catalog": "{\"CFSVolumes\":[],\"COSVolumes\":[{\"Bucket\":\"aidanyxu-cos-260200065\",\"VolumeSubPath\":\"\",\"SubPathMode\":\"subPath\",\"SubPath\":\"\",\"MountPath\":\"/\",\"VolumeMountMode\":\"ReadWrite\",\"Source\":\"checkpoint\"}],\"CFSTurboVolumes\":[],\"GooseFSVolumes\":[]}",
"MaxKeep": 0,
"OutputDir": "",
"SaveFreq": 1,
"SaveStrategy": "epoch"
},
"CheckpointMountInfo": {
"Bucket": "aidanyxu-cos-260200065",
"MountPath": "/",
"PlatformManaged": false,
"Region": "ap-guangzhou",
"StorageType": "COS",
"VolumeSubPath": ""
},
"CreateTime": 1784893623242,
"Creator": "700002655693",
"Description": "sadfafda",
"Entrypoint": "python /workspace/ray_post_training_entry.py --num-workers 1 --gpus-per-worker 0 --cpus-per-worker 1",
"HasRunningInstances": false,
"InstanceCount": 0,
"Kind": "POST_TRAINING",
"MlFlowConfig": "{\"MlFlowMode\":\"local\",\"MlFlowServerOptions\":{\"cors-allowed-origins\":\"*\"}}",
"Mode": "sft",
"OutputModelName": "",
"Priority": 5,
"Queue": "default",
"ResourceConfig": "{\"Head\":{\"Name\":\"Head\",\"ResourceType\":\"CPU\",\"BillingItem\":\"\",\"Spec\":2,\"Envs\":[],\"Labels\":[],\"ResourcesLabels\":[],\"HighAvailability\":false,\"PodNum\":1},\"Worker\":[{\"Name\":\"WorkerGroup01\",\"ResourceType\":\"CPU\",\"BillingItem\":\"\",\"Spec\":1,\"Envs\":[],\"Labels\":[],\"ResourcesLabels\":[],\"MinPodNum\":1,\"MaxPodNum\":1}]}",
"ResourceConfigId": "",
"ResourcePartitionId": "dlc-p-wdtiljwu",
"ResourcePartitionName": "leion",
"RuntimeEnv": "{\"env_vars\":{\"TRAIN_CONFIG_YAML_B64\":\"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\",\"DEEPSPEED_CONFIG_B64\":\"ewogICAgInRyYWluX2JhdGNoX3NpemUiOiAiYXV0byIsCiAgICAidHJhaW5fbWljcm9fYmF0Y2hfc2l6ZV9wZXJfZ3B1IjogImF1dG8iLAogICAgImJmMTYiOiB7ImVuYWJsZWQiOiAiYXV0byJ9LAogICAgImdyYWRpZW50X2NsaXBwaW5nIjogImF1dG8iLAogICAgInplcm9fb3B0aW1pemF0aW9uIjogewogICAgICAgICJvZmZsb2FkX29wdGltaXplciI6IHsiZGV2aWNlIjogIm5vbmUifSwKICAgICAgICAib3ZlcmxhcF9jb21tIjogdHJ1ZSwKICAgICAgICAic3RhZ2UiOiAzLAogICAgICAgICJzdWJfZ3JvdXBfc2l6ZSI6IDFFKzksCiAgICAgICAgImNvbnRpZ3VvdXNfZ3JhZGllbnRzIjogdHJ1ZSwKICAgICAgICAicmVkdWNlX2J1Y2tldF9zaXplIjogImF1dG8iLAogICAgICAgICJzdGFnZTNfZ2F0aGVyXzE2Yml0X3dlaWdodHNfb25fbW9kZWxfc2F2ZSI6IHRydWUsCiAgICAgICAgIm9mZmxvYWRfcGFyYW0iOiB7ImRldmljZSI6ICJub25lIn0sCiAgICAgICAgInN0YWdlM19wcmVmZXRjaF9idWNrZXRfc2l6ZSI6ICJhdXRvIiwKICAgICAgICAic3RhZ2UzX3BhcmFtX3BlcnNpc3RlbmNlX3RocmVzaG9sZCI6ICJhdXRvIiwKICAgICAgICAic3RhZ2UzX21heF9saXZlX3BhcmFtZXRlcnMiOiAxRSs5LAogICAgICAgICJzdGFnZTNfbWF4X3JldXNlX2Rpc3RhbmNlIjogMUUrOQogICAgfSwKICAgICJ6ZXJvX2FsbG93X3VudGVzdGVkX29wdGltaXplciI6IHRydWUsCiAgICAiZ3JhZGllbnRfYWNjdW11bGF0aW9uX3N0ZXBzIjogImF1dG8iLAogICAgImZwMTYiOiB7CiAgICAgICAgIm1pbl9sb3NzX3NjYWxlIjogMSwKICAgICAgICAiaW5pdGlhbF9zY2FsZV9wb3dlciI6IDE2LAogICAgICAgICJsb3NzX3NjYWxlIjogMCwKICAgICAgICAiZW5hYmxlZCI6ICJhdXRvIiwKICAgICAgICAibG9zc19zY2FsZV93aW5kb3ciOiAxMDAwLAogICAgICAgICJoeXN0ZXJlc2lzIjogMgogICAgfQp9Cg==\",\"NUM_WORKERS\":\"1\",\"GPUS_PER_WORKER\":\"0\",\"CPUS_PER_WORKER\":\"1\",\"NCCL_DEBUG\":\"WARN\",\"HF_HOME\":\"/workspace/cache/hf\",\"WANDB_DISABLED\":\"true\",\"MLFLOW_TRACKING_URI\":\"$MLFLOW_TRACKING_URI\",\"MLFLOW_EXPERIMENT_NAME\":\"xxd\",\"HF_MLFLOW_LOG_ARTIFACTS\":\"false\"}}",
"SpecId": "raytrain-spec-tiogqf-yb6e",
"SpecName": "xxd",
"Tags": [],
"TuningParams": {
"CutoffLen": 4096,
"DPOBeta": 0.1,
"DPOLoss": "sigmoid",
"Epochs": 3,
"FineTuneType": "lora",
"GradientAccumulationSteps": 4,
"GradientCheckPointing": false,
"LoraAlpha": 16,
"LoraDropout": 0.05,
"LoraRank": 8,
"LoraTarget": "all",
"LrScheduler": "cosine",
"PerDeviceBatchSize": 2,
"TrainingMode": "balanced",
"WarmupRatio": 0.03
},
"UpdateTime": 1784893623242
},
"RequestId": "72056b91-6068-4df5-a39e-a24fc8512bc6"
}
}5. 开发者资源
腾讯云 API 平台
腾讯云 API 平台 是综合 API 文档、错误码、API Explorer 及 SDK 等资源的统一查询平台，方便您从同一入口查询及使用腾讯云提供的所有 API 服务。
API Inspector
用户可通过 API Inspector 查看控制台每一步操作关联的 API 调用情况，并自动生成各语言版本的 API 代码，也可前往 API Explorer 进行在线调试。
SDK
云 API 3.0 提供了配套的开发工具集（SDK），支持多种编程语言，能更方便的调用 API。
- Tencent Cloud SDK 3.0 for Python: CNB, GitHub, Gitee
- Tencent Cloud SDK 3.0 for Java: CNB, GitHub, Gitee
- Tencent Cloud SDK 3.0 for PHP: CNB, GitHub, Gitee
- Tencent Cloud SDK 3.0 for Go: CNB, GitHub, Gitee
- Tencent Cloud SDK 3.0 for Node.js: CNB, GitHub, Gitee
- Tencent Cloud SDK 3.0 for .NET: CNB, GitHub, Gitee
- Tencent Cloud SDK 3.0 for C++: CNB, GitHub, Gitee
- Tencent Cloud SDK 3.0 for Ruby: CNB, GitHub, Gitee
命令行工具
6. 错误码
该接口暂无业务逻辑相关的错误码，其他错误码详见 公共错误码。
