Skip to content
Back to skills

Firered Image Edit V1 1

ASecurity

Firered Image | Edit | v1.1. Edit images with text instructions, multi-image reference, and configurable quality settings. Triggers: image edit, firered, edit image, inpaint, modify

  • 29 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added June 2, 2026
documentationbashapidocumentation

Works with

  • api

Security analysis

A100/100

Scanned June 2, 2026

npx -y skills add awesome-genmedia/skills --skill firered-image-edit-v1-1 --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Firered Image Edit V1 1?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Firered Image Edit V1 1
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/awesome-genmedia-firered-image-edit-v1-1/badge)](https://www.skillsdirectory.com/skills/awesome-genmedia-firered-image-edit-v1-1)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: firered-image-edit-v1-1
description: "Firered Image | Edit | v1.1. Edit images with text instructions, multi-image reference, and configurable quality settings. Triggers: image edit, firered, edit image, inpaint, modify"
allowed-tools: Bash(curl *), WebFetch
---

# Firered Image | Edit | v1.1

Edit images using text-based instructions with support for multi-image reference, negative prompts, configurable inference steps, and multiple output formats.

## Quick Start

> Requires an each::labs API key. Get one at [eachlabs.ai](https://eachlabs.ai).

```bash
curl -X POST https://api.eachlabs.ai/v1/prediction \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -d '{
    "model": "firered-image-edit-v1-1",
    "version": "0.0.1",
    "input": {
      "prompt": "Change the sky to a dramatic sunset with orange and purple clouds",
      "image_urls": ["https://example.com/landscape.jpg"],
      "image_size": "landscape_16_9",
      "num_inference_steps": 30,
      "guidance_scale": 4
    }
  }'
```

## Parameters

| Parameter | Type | Default | Description |
|-----------|------|---------|-------------|
| acceleration | string | regular | Inference speed optimization level. enum: none, regular, high |
| enable_safety_checker | boolean | true | Enable the safety checker for generated images. |
| guidance_scale | number | 4 | Classifier-free guidance scale. Higher values follow the prompt more closely. |
| image_size | string | square_hd | Size of the generated image. If None, uses input image dimensions. enum: square_hd, square, portrait_4_3, portrait_16_9, landscape_4_3, landscape_16_9 |
| image_urls | array | | URLs of the images to edit. Supports single and multi-image reference. |
| negative_prompt | string | | Negative prompt to exclude unwanted content. |
| num_images | integer | 1 | Number of images to generate. |
| num_inference_steps | integer | 30 | Number of inference steps. More steps = higher quality but slower. |
| output_format | string | png | Output image format. enum: jpeg, png |
| prompt | string | | Editing instruction describing what changes to make. |
| seed | string | | Random seed for reproducible results. |

## Examples

**Edit a landscape photo:**
```bash
curl -X POST https://api.eachlabs.ai/v1/prediction \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -d '{
    "model": "firered-image-edit-v1-1",
    "version": "0.0.1",
    "input": {
      "prompt": "Add a rainbow arching across the sky",
      "image_urls": ["https://example.com/countryside.jpg"],
      "image_size": "landscape_16_9",
      "guidance_scale": 5,
      "num_inference_steps": 40
    }
  }'
```

**Generate multiple variations:**
```bash
curl -X POST https://api.eachlabs.ai/v1/prediction \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -d '{
    "model": "firered-image-edit-v1-1",
    "version": "0.0.1",
    "input": {
      "prompt": "Transform the room interior to a modern minimalist style",
      "image_urls": ["https://example.com/room.jpg"],
      "negative_prompt": "cluttered, messy, dark",
      "num_images": 3,
      "output_format": "jpeg"
    }
  }'
```

**Fast edit with high acceleration:**
```bash
curl -X POST https://api.eachlabs.ai/v1/prediction \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -d '{
    "model": "firered-image-edit-v1-1",
    "version": "0.0.1",
    "input": {
      "prompt": "Replace the car with a bicycle",
      "image_urls": ["https://example.com/street.jpg"],
      "acceleration": "high",
      "num_inference_steps": 15,
      "seed": "42"
    }
  }'
```

## Related Models

- [p-image-edit-lora-image-edit](../p-image-edit-lora-image-edit/) - Image editing with LoRA weights
- [nano-banana-2-edit](../nano-banana-2-edit/) - Nano Banana image editing
- [xai-grok-imagine-image-edit](../xai-grok-imagine-image-edit/) - Grok image editing

## Documentation

- [each::labs Docs](https://docs.eachlabs.ai)
- [API Reference](https://docs.eachlabs.ai/api/overview)

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.