higgsfield-soul-id
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
ChineseHiggsfield Soul Character
Higgsfield Soul Character
Train a face-faithful identity model. Reusable across all Soul-powered generations.
训练忠实还原人脸特征的身份模型。可在所有基于Soul的生成任务中重复使用。
Step 0 — Bootstrap
步骤0 — 准备工作
Before any other command:
- If is not on
higgsfield, install it:$PATHbashcurl -fsSL https://raw.githubusercontent.com/higgsfield-ai/cli/main/install.sh | sh - If fails with
higgsfield account status/Session expired, ask the user to runNot authenticated(interactive) and wait for confirmation.higgsfield auth login - Soul training requires a paid plan (Basic+). If shows free plan, tell the user before submitting.
higgsfield account status
在执行任何其他命令前:
- 如果不在
higgsfield中,请安装它:$PATHbashcurl -fsSL https://raw.githubusercontent.com/higgsfield-ai/cli/main/install.sh | sh - 如果返回
higgsfield account status(会话过期)或Session expired(未认证),请让用户运行Not authenticated(交互式登录)并等待确认。higgsfield auth login - Soul训练需要付费套餐(基础版及以上)。如果显示为免费套餐,请在提交前告知用户。
higgsfield account status
UX Rules
用户体验规则
- Be concise. No raw IDs in chat. Just say "Soul ready" with a name reference.
- Detect language and respond in it. CLI flags stay English.
- Ask for the smallest set of inputs: name + photos. Pick a sensible model variant.
- Polling is silent — training takes minutes. Don't repeat status updates.
- 保持简洁。聊天中不要显示原始ID,只需告知“Soul已准备就绪”并附上名称标识。
- 检测用户语言并使用对应语言回复。CLI参数保持英文。
- 仅请求必要的最少输入:名称+照片。选择合适的模型变体。
- 轮询过程静默进行——训练需要数分钟。不要重复发送状态更新。
Workflow
工作流程
- Get name. One word, used for later reference. Ask if missing.
- Get photos. 5–20 face photos, varied angles and lighting. Local paths or already-uploaded IDs both work — accepts either.
--image - Pick variant.
- — for image generation (default)
--soul-2 - — for cinematic / video work Choose based on user's stated downstream use. Default to
--soul-cinematic.--soul-2
- Submit.
CLI auto-uploads paths. Captures returned reference id.bash
higgsfield soul-id create --name "<name>" --soul-2 --image ./photo1.png --image ./photo2.png ... higgsfield soul-id create --name "<name>" --soul-2 --image <upload_id> --image <upload_id> ... - Wait. . Silent. Default timeout 30m.
higgsfield soul-id wait <id> - Deliver. "Soul ready. Use in generate with
<name>."--soul-id <id>
- 获取名称:单个单词,用于后续标识。如果缺失请询问用户。
- 获取照片:5-20张人脸照片,包含不同角度和光线条件。本地路径或已上传的ID均可——参数支持两种格式。
--image - 选择模型变体:
- — 用于图像生成(默认选项)
--soul-2 - — 用于电影级/视频制作 根据用户说明的后续用途选择,默认使用
--soul-cinematic。--soul-2
- 提交训练:
CLI会自动上传本地路径的文件。捕获返回的reference_id。bash
higgsfield soul-id create --name "<name>" --soul-2 --image ./photo1.png --image ./photo2.png ... higgsfield soul-id create --name "<name>" --soul-2 --image <upload_id> --image <upload_id> ... - 等待完成:。静默执行,默认超时时间30分钟。
higgsfield soul-id wait <id> - 交付结果:“Soul 已准备就绪。请在生成任务中使用
<name>参数。”--soul-id <id>
Use the Soul
使用Soul模型
Once trained, pass to :
higgsfield-generatebash
higgsfield generate create text2image_soul_v2 --prompt "..." --soul-id <ref_id> --wait
higgsfield generate create soul_cinema_studio --prompt "..." --soul-id <ref_id> --wait训练完成后,将其传入:
higgsfield-generatebash
higgsfield generate create text2image_soul_v2 --prompt "..." --soul-id <ref_id> --wait
higgsfield generate create soul_cinema_studio --prompt "..." --soul-id <ref_id> --waitListing existing Souls
列出已有的Soul模型
bash
higgsfield soul-id list # all references
higgsfield soul-id get <id> # one by idbash
higgsfield soul-id list # 列出所有标识
higgsfield soul-id get <id> # 根据ID查询单个标识Errors
错误处理
- — user is on free plan; tell them.
Minimum Basic plan required - — check photos quality (5+ unique faces, well-lit).
Training failed - →
Session expired.higgsfield auth login
- — 用户当前为免费套餐,请告知用户。
Minimum Basic plan required - — 检查照片质量(需5张以上不同角度的清晰人脸,光线充足)。
Training failed - → 执行
Session expired重新登录。higgsfield auth login
Reference docs
参考文档
- — what photos work best
references/photo-guide.md - — common training failures
references/troubleshooting.md
- — 最佳照片指南
references/photo-guide.md - — 常见训练失败问题排查
references/troubleshooting.md