列出训练运行 - Fastino API Docs

GET

https://api.fastino.ai

training

jobs

curl "https://api.fastino.ai/v1/training/jobs?type=interactive&limit=20&offset=0" \
  -H "X-API-Key: $FASTINO_API_KEY"

{
  "success": true,
  "training_jobs": [
    {
      "type": "interactive",
      "training_run_id": "YOUR_SESSION_ID:train:0",
      "base_model": "Qwen/Qwen3.6-35B-A3B",
      "model_owner": "YOUR_TEAM_ID",
      "is_lora": true,
      "corrupted": false,
      "lora_rank": 32,
      "last_request_time": "2026-10-09T18:42:10Z",
      "last_checkpoint": null,
      "last_sampler_checkpoint": {
        "checkpoint_id": "sampler_weights/step-100",
        "checkpoint_type": "sampler",
        "time": "2026-10-09T18:40:02Z",
        "tinker_path": "tinker://YOUR_SESSION_ID:train:0/sampler_weights/step-100",
        "size_bytes": 167772160,
        "public": false,
        "user_metadata": null,
        "expires_at": null
      },
      "user_metadata": {"experiment": "baseline"}
    }
  ],
  "count": 1,
  "total": 1,
  "has_more": false
}

GET /v1/training/jobs 按从新到旧的顺序返回已认证调用方可见的训练运行。训练运行是您通过兼容 Tinker 的 SDK 创建并逐步驱动的训练。将 SDK 的 base_url 设置为 https://api.fastino.ai/v1/training/compat/tinker，并通过 SDK 的凭据命令传入您的 Fastino API 密钥：TINKER_CREDENTIAL_CMD='printf %s "$FASTINO_API_KEY"'。Tinker 的 tml- 密钥会被拒绝并返回 401。您可以看到自己的运行，以及团队在您有权访问的项目中的运行。

请求

string

必填

您的 Fastino API 密钥。

string

作业类型。使用 interactive（此处的默认值）。config 返回 400。

integer

默认值:"20"

每页大小。取值会被限制在 1 到 1000 之间。

integer

默认值:"0"

要跳过的运行数量。

curl "https://api.fastino.ai/v1/training/jobs?type=interactive&limit=20&offset=0" \
  -H "X-API-Key: $FASTINO_API_KEY"

响应

boolean

请求是否成功。

object[]

本页的训练运行。每个运行包含获取训练运行中描述的字段。

integer

本页中的运行数量。

integer

分页前可见运行的总数。

boolean

此页之后是否还有下一页。

{
  "success": true,
  "training_jobs": [
    {
      "type": "interactive",
      "training_run_id": "YOUR_SESSION_ID:train:0",
      "base_model": "Qwen/Qwen3.6-35B-A3B",
      "model_owner": "YOUR_TEAM_ID",
      "is_lora": true,
      "corrupted": false,
      "lora_rank": 32,
      "last_request_time": "2026-10-09T18:42:10Z",
      "last_checkpoint": null,
      "last_sampler_checkpoint": {
        "checkpoint_id": "sampler_weights/step-100",
        "checkpoint_type": "sampler",
        "time": "2026-10-09T18:40:02Z",
        "tinker_path": "tinker://YOUR_SESSION_ID:train:0/sampler_weights/step-100",
        "size_bytes": 167772160,
        "public": false,
        "user_metadata": null,
        "expires_at": null
      },
      "user_metadata": {"experiment": "baseline"}
    }
  ],
  "count": 1,
  "total": 1,
  "has_more": false
}

无效的分页值会返回 422。缺少 API 密钥或密钥无效会返回 401。如果您的团队未启用训练运行，该路由会返回 404。

⌘I

curl "https://api.fastino.ai/v1/training/jobs?type=interactive&limit=20&offset=0" \
  -H "X-API-Key: $FASTINO_API_KEY"

{
  "success": true,
  "training_jobs": [
    {
      "type": "interactive",
      "training_run_id": "YOUR_SESSION_ID:train:0",
      "base_model": "Qwen/Qwen3.6-35B-A3B",
      "model_owner": "YOUR_TEAM_ID",
      "is_lora": true,
      "corrupted": false,
      "lora_rank": 32,
      "last_request_time": "2026-10-09T18:42:10Z",
      "last_checkpoint": null,
      "last_sampler_checkpoint": {
        "checkpoint_id": "sampler_weights/step-100",
        "checkpoint_type": "sampler",
        "time": "2026-10-09T18:40:02Z",
        "tinker_path": "tinker://YOUR_SESSION_ID:train:0/sampler_weights/step-100",
        "size_bytes": 167772160,
        "public": false,
        "user_metadata": null,
        "expires_at": null
      },
      "user_metadata": {"experiment": "baseline"}
    }
  ],
  "count": 1,
  "total": 1,
  "has_more": false
}

​请求

​响应

请求

响应