About openrouter-img
A command-line skill for generating and editing images through OpenRouter's Image API. It supports model aliases, resolution and aspect-ratio options, and can list models with current endpoint prices.
Key features
- 01Generate images through OpenRouter
- 02Edit an image from an input file
- 03Select model, resolution, and ratio
- 04List models and endpoint prices
- 05Read API key from environment files
Use cases
- 01Generate an image from a prompt
- 02Edit an existing image
- 03Compare available image models
- 04Check endpoint prices before generation
Install
Copy the command and run it in a terminal.
npx skills add https:/After installing, type/openrouter-imgin Claude Code to confirm it loaded.
Bundled filesagents/openai.yamlscripts/generate_image.py
Details
From the project READMEOpenRouter Img
Gera imagens pela Image API do OpenRouter (POST /api/v1/images). Chave: OPENROUTER_API_KEY (cwd .env ou ~/.env).
Uso
# Listar top 10 + preços ao vivo do endpoint
uv run ~/.agents/skills/openrouter-img/scripts/generate_image.py --list --top
# Gerar
uv run ~/.agents/skills/openrouter-img/scripts/generate_image.py \
--prompt "..." --filename "yyyy-mm-dd-hh-mm-ss-name.png" \
--model flash --resolution 1K --aspect-ratio 16:9
Aliases de --model: ver tabela abaixo — ou o slug completo.
Default: flash @ 1K / 16:9. Só subir resolução se o usuário pedir. Rodar no cwd do usuário. Não ler a imagem de volta — informar path (+ usage.cost se vier).
Edição: --input-image path.png + prompt de edição.
Modelos empregados (top 10)
| # | Alias | Nome | Slug OpenRouter |
|---|---|---|---|
| 1 | pro / nb-pro | Nano Banana Pro (Gemini 3 Pro Image) | google/gemini-3-pro-image |
| 2 | gpt / gpt2 | GPT Image 2 | openai/gpt-image-2 |
| 3 | flux-max | FLUX.2 Max | black-forest-labs/flux.2-max |
| 4 | flux | FLUX.2 Pro | black-forest-labs/flux.2-pro |
| 5 | seedream-pro | Seedream 5.0 Pro | bytedance-seed/seedream-5-0-pro |
| 6 | seedream | Seedream 5.0 Lite | bytedance-seed/seedream-5-0-lite |
| 7 | flash / nb-flash | Nano Banana 2 (Gemini 3.1 Flash Image) | google/gemini-3.1-flash-image |
| 8 | qwen | Qwen Image 3 Pro | qwen/qwen-image-3-pro |
| 9 | grok | Grok Imagine Image 2.0 | x-ai/grok-imagine-image-2.0 |
| 10 | seedream-4.5 | Seedream 4.5 | bytedance-seed/seedream-4.5 |
Tabela de custo (consulta)
Valores OpenRouter (snapshot ago/2026). Atualizar com --list --top. O response usage.cost é a verdade da cobrança.
| Alias | 512 | 1K | 2K | 4K / outro | Unidade |
|---|---|---|---|---|---|
flash | ~$0.045 | ~$0.067 | ~$0.101 | ~$0.151 @4K | token → $/img estimado |
pro | — | ~$0.134 | ~$0.134 | — (sem 4K neste endpoint) | token → $/img estimado |
seedream | — | — | $0.035 | $0.035 @4K | $/imagem fixo |
seedream-pro | — | $0.045 | $0.045 | $0.09 high-res | $/imagem |
seedream-4.5 | — | $0.04 | $0.04 | $0.04 | $/imagem |
qwen | — | $0.04 | $0.075 | — | $/imagem |
grok | — | $0.04–0.06 | $0.06–0.08 | — | $/imagem (low/medium) |
flux | — | ~$0.03 | ~$0.12* | — | $0.03 / megapixel |
flux-max | — | ~$0.07 | ~$0.28* | — | $0.07 / megapixel |
gpt2 | — | ver usage.cost | ver usage.cost | — | token (input+output) |
*Estimativa 2K ≈ 4 MP (2048²); o custo real depende dos pixels gerados.
Resoluções “—” = não listadas / não suportadas nesse endpoint. Se a API rejeitar, mudar --resolution.
Escolha rápida
- UI/mock com texto →
proouflash - Volume barato →
flash@512,seedream,flux - Fotorrealismo / arte →
flux-max,seedream-pro,gpt2