Workflow Examples

Complete, copy-paste ready workflows based on production use cases from the Jetty platform.

Quick Reference

Example	Use Case	Complexity
Hello World	First workflow	Beginner
LLM Chat	Basic LLM interaction	Beginner
Text-to-Image	Image generation	Beginner
Image Evaluation	Judge generated images	Intermediate
Batch Processing	Process multiple items	Intermediate
Multi-Model Comparison	A/B testing	Advanced
Document Translation	Process documents	Advanced

Hello World

Simplest possible workflow using text_doubler.

Based on: hello_flow

{
  "init_params": {
    "text": "Hello, Jetty!"
  },
  "step_configs": {
    "doubler": {
      "activity": "text_doubler",
      "text_path": "init_params.text"
    }
  },
  "steps": ["doubler"]
}

Output: "Hello, Jetty!Hello, Jetty!"

LLM Chat

Basic LLM interaction with any provider.

Based on: litellm-chat

{
  "init_params": {
    "user_question": "What are the key differences between Python and JavaScript?",
    "system_message": "You are a helpful programming tutor."
  },
  "step_configs": {
    "chat": {
      "model": "gemini/gemini-1.5-flash",
      "activity": "litellm_chat",
      "prompt": "init_params.user_question",
      "max_tokens": 500,
      "temperature": 0.7,
      "system_prompt": "init_params.system_message"
    }
  },
  "steps": ["chat"]
}

Text-to-Image

Generate images from text prompts.

Based on: text2image_with_metadata

{
  "init_params": {
    "prompt": "A beautiful landscape with mountains and a lake"
  },
  "step_configs": {
    "generate": {
      "model": "black-forest-labs/flux-schnell",
      "activity": "replicate_text2image",
      "prompt_path": "init_params.prompt",
      "output_format": "jpg"
    },
    "add_metadata": {
      "activity": "add_image_metadata"
    }
  },
  "steps": ["generate", "add_metadata"]
}

Image Evaluation

Generate and evaluate images with multiple criteria.

Based on: llm-judge

{
  "init_params": {
    "prompt": "A cow eating a shrimp",
    "model": "black-forest-labs/flux-schnell",
    "evaluation_criteria": "Are there animals in this picture?"
  },
  "step_configs": {
    "generate": {
      "activity": "replicate_text2image",
      "model_path": "init_params.model",
      "prompt_path": "init_params.prompt",
      "aspect_ratio": "16:9",
      "output_format": "jpg"
    },
    "describe": {
      "model": "gpt-4o",
      "activity": "simple_judge",
      "items_path": "generate.outputs.images[0].path",
      "judge_type": "scale",
      "scale_range": [1, 5],
      "model_provider": "openai",
      "instruction_path": "init_params.evaluation_criteria"
    },
    "ip_risk": {
      "model": "gpt-4o",
      "activity": "simple_judge",
      "items_path": "generate.outputs.images[0].path",
      "judge_type": "scale",
      "instruction": "Rate the potential for IP infringement in this image.",
      "scale_range": [0, 1],
      "model_provider": "openai"
    }
  },
  "steps": ["generate", "describe", "ip_risk"]
}

Batch Processing

Process multiple items in parallel with fan-out/fan-in.

Based on: emit-doubler-collector

{
  "init_params": {
    "start": 1,
    "end": 5
  },
  "step_configs": {
    "generate_numbers": {
      "mode": "range",
      "step": 1,
      "activity": "number_sequence_generator",
      "end_path": "init_params.end",
      "start_path": "init_params.start"
    },
    "process_each": {
      "activity": "list_emit_await",
      "items_path": "generate_numbers.outputs.sequence",
      "task_reference": {
        "task_name": "hello_flow"
      },
      "data_mapping": {
        "text": "Processing number {{ $item }}"
      }
    },
    "collect": {
      "activity": "extract_from_trajectories",
      "trajectory_list_path": "process_each.outputs.trajectory_references",
      "extract_keys": {
        "result": "doubler.outputs.text"
      }
    }
  },
  "steps": ["generate_numbers", "process_each", "collect"]
}

Multi-Model Comparison

Compare outputs from different image models.

Based on: t2i-evals

{
  "init_params": {
    "prompt": "A beautiful landscape with mountains and a lake"
  },
  "step_configs": {
    "flux": {
      "model": "black-forest-labs/flux-schnell",
      "activity": "replicate_text2image",
      "prompt_path": "init_params.prompt"
    },
    "midjourney": {
      "model": "tstramer/midjourney-diffusion:436b051ebd8f68d23e83d22de5e198e0995357afef113768c20f0b6fcef23c8b",
      "activity": "replicate_text2image",
      "prompt_path": "init_params.prompt",
      "guidance_scale": 7.5,
      "num_inference_steps": 50
    }
  },
  "steps": ["flux", "midjourney"]
}

Document Translation

Translate documents with validation.

Based on: sacred-translate-v1

{
  "init_params": {
    "target_language": "french"
  },
  "step_configs": {
    "translate": {
      "model": "gemini-2.5-pro",
      "activity": "gemini_file_reader",
      "asset_path": "init_params.file_paths[0]",
      "prompt": "Translate this document. Preserve formatting."
    },
    "save": {
      "activity": "save_text_file",
      "file_text_path": "translate.outputs.text"
    },
    "read_original": {
      "activity": "read_text_file",
      "text_path": "init_params.file_paths[0]"
    },
    "combine": {
      "activity": "text_concatenate",
      "text_paths": [
        "translate.outputs.text",
        "read_original.outputs.text"
      ]
    },
    "validate": {
      "model": "gemini-2.5-pro",
      "activity": "gemini_text_reader",
      "text_path": "combine.outputs.json",
      "prompt": "Compare translation with original. List any issues."
    }
  },
  "steps": ["translate", "save", "read_original", "combine", "validate"]
}

Image to Video

Generate video from text-to-image.

Based on: text2image2video

{
  "init_params": {
    "prompt": "A scientist holding up a balloon",
    "motion": "zoom out to space"
  },
  "step_configs": {
    "expand": {
      "model": "openai/gpt-4",
      "activity": "litellm_chat",
      "messages_path": "init_params.prompt",
      "system_prompt": "Elaborate the prompt with cyberpunk themes."
    },
    "generate_image": {
      "model": "black-forest-labs/flux-schnell",
      "activity": "replicate_text2image",
      "prompt_path": "expand.outputs.text",
      "output_format": "jpg"
    },
    "animate": {
      "activity": "replicate_text2video",
      "duration": 3,
      "image_path": "generate_image.outputs.images[0].path",
      "resolution": "480p",
      "prompt_path": "init_params.motion"
    }
  },
  "steps": ["expand", "generate_image", "animate"]
}

Running Examples

All examples can be run via the API:

# Sync execution (wait for result)
curl -X POST "https://flows-api.jetty.io/api/v1/run-sync/YOUR_COLLECTION/TASK_NAME" \
  -H "Authorization: Bearer $JETTY_API_TOKEN" \
  -F "bakery_host=https://dock.jetty.io" \
  -F 'init_params={"prompt": "your prompt here"}'

# Async execution (get tracking ID)
curl -X POST "https://flows-api.jetty.io/api/v1/run/YOUR_COLLECTION/TASK_NAME" \
  -H "Authorization: Bearer $JETTY_API_TOKEN" \
  -F "bakery_host=https://dock.jetty.io" \
  -F 'init_params={"prompt": "your prompt here"}'

Quick Reference​

Hello World​

LLM Chat​

Text-to-Image​

Image Evaluation​

Batch Processing​

Multi-Model Comparison​

Document Translation​

Image to Video​

Running Examples​