gaz-mcp image generation
Generate images through the configured gaz-mcp MCP server. The server uses a closed model allowlist; never invent a model ID or assume that every OpenRouter image model is enabled.
Start safely
- Call
image_model_list and use only the exact returned model IDs.
- Confirm the requested model, prompt, dimensions, format, and number of outputs before expensive generations.
- Call
image_generate once per requested model. Generation is non-idempotent and may incur provider cost; do not retry automatically after an ambiguous timeout.
- Return or inspect the MCP
image content blocks. Structured metadata identifies the model, MIME type, creation time, and usage/cost when supplied.
Tool contracts
image_model_list
Returns the configured allowlist. Use it as the source of truth for model selection and comparisons.
image_generate
Required arguments:
| Argument |
Meaning |
model |
Exact ID from image_model_list. |
prompt |
Non-empty description of the desired image. |
Optional arguments:
| Argument |
Meaning |
input_references |
Up to 10 seed images. Each item is either an HTTP(S) url, or base64 data plus an image media_type such as image/png. |
n |
1–10 images; default 1. Provider support varies. |
size |
Exact pixels or a tier such as 2K; do not combine with resolution or aspect_ratio. |
resolution, aspect_ratio |
Normalized dimensions such as 2K and 16:9; support varies by model. |
quality |
auto, low, medium, or high, when supported. |
output_format |
png, jpeg, webp, or svg, when supported. |
background |
auto, transparent, or opaque, when supported. |
output_compression |
0–100 for JPEG/WebP where supported. |
seed |
Integer provider seed where supported. |
provider |
OpenRouter routing: only, order, ignore, sort, allow_fallbacks, and provider-specific options. |
Example request:
image_generate(
model="bytedance-seed/seedream-5-0-lite",
prompt="Editorial photograph of a red bicycle beside a canal at sunrise",
size="2048x2048",
quality="high",
n=1
)
Model limits are enforced by the provider. If rejected, preserve the prompt and adjust only the reported incompatible option; some models require a minimum pixel count or do not support a requested format.
Reference images
Use a public HTTPS URL when the provider can fetch it. For local bytes, send valid standard base64 and an image/* MIME type:
{"input_references":[{"data":"<base64>","media_type":"image/png"}]}
Do not place secrets in prompts, URLs, provider options, or reference data. Do not expose API keys or reproduce large base64 payloads in prose.
Compare models
- Call
image_model_list.
- Select two or more exact IDs.
- Send equivalent
prompt, references, dimensions, quality, and seed to one image_generate call per model.
- Compare returned image blocks and report model IDs, option differences, provider errors, and usage/cost separately from subjective visual judgment.
If no models are configured, the image tools are not registered; report the configuration prerequisite instead of guessing a fallback model.
1---2name: gaz-mcp-image-generation3description: Use the gaz-mcp OpenRouter image-generation proxy to list configured models, generate images from prompts and reference images, and compare outputs across models. Use when an MCP server exposes image_model_list or image_generate and the task requires creating, iterating, or evaluating images.4---56# gaz-mcp image generation78Generate images through the configured `gaz-mcp` MCP server. The server uses a closed model allowlist; never invent a model ID or assume that every OpenRouter image model is enabled.910## Start safely11121. Call `image_model_list` and use only the exact returned model IDs.132. Confirm the requested model, prompt, dimensions, format, and number of outputs before expensive generations.143. Call `image_generate` once per requested model. Generation is non-idempotent and may incur provider cost; do not retry automatically after an ambiguous timeout.154. Return or inspect the MCP `image` content blocks. Structured metadata identifies the model, MIME type, creation time, and usage/cost when supplied.1617## Tool contracts1819### `image_model_list`2021Returns the configured allowlist. Use it as the source of truth for model selection and comparisons.2223### `image_generate`2425Required arguments:2627| Argument | Meaning |28|---|---|29| `model` | Exact ID from `image_model_list`. |30| `prompt` | Non-empty description of the desired image. |3132Optional arguments:3334| Argument | Meaning |35|---|---|36| `input_references` | Up to 10 seed images. Each item is either an HTTP(S) `url`, or base64 `data` plus an image `media_type` such as `image/png`. |37| `n` | 1–10 images; default 1. Provider support varies. |38| `size` | Exact pixels or a tier such as `2K`; do not combine with `resolution` or `aspect_ratio`. |39| `resolution`, `aspect_ratio` | Normalized dimensions such as `2K` and `16:9`; support varies by model. |40| `quality` | `auto`, `low`, `medium`, or `high`, when supported. |41| `output_format` | `png`, `jpeg`, `webp`, or `svg`, when supported. |42| `background` | `auto`, `transparent`, or `opaque`, when supported. |43| `output_compression` | 0–100 for JPEG/WebP where supported. |44| `seed` | Integer provider seed where supported. |45| `provider` | OpenRouter routing: `only`, `order`, `ignore`, `sort`, `allow_fallbacks`, and provider-specific `options`. |4647Example request:4849```text50image_generate(51 model="bytedance-seed/seedream-5-0-lite",52 prompt="Editorial photograph of a red bicycle beside a canal at sunrise",53 size="2048x2048",54 quality="high",55 n=156)57```5859Model limits are enforced by the provider. If rejected, preserve the prompt and adjust only the reported incompatible option; some models require a minimum pixel count or do not support a requested format.6061## Reference images6263Use a public HTTPS URL when the provider can fetch it. For local bytes, send valid standard base64 and an `image/*` MIME type:6465```json66{"input_references":[{"data":"<base64>","media_type":"image/png"}]}67```6869Do not place secrets in prompts, URLs, provider options, or reference data. Do not expose API keys or reproduce large base64 payloads in prose.7071## Compare models72731. Call `image_model_list`.742. Select two or more exact IDs.753. Send equivalent `prompt`, references, dimensions, quality, and seed to one `image_generate` call per model.764. Compare returned image blocks and report model IDs, option differences, provider errors, and usage/cost separately from subjective visual judgment.7778If no models are configured, the image tools are not registered; report the configuration prerequisite instead of guessing a fallback model.
jcastilloa