Skip to main content
Glama

AdsTurbo

by io.github.AdsTurbo
OverviewDiscussionsAdmin

image_create

图片生成 / 编辑

每次调用固定生成 1 张。 默认异步：返回 workspace_id（status=pending，result_type=image），用 work/status 轮询取 result_url； sync_mod=true 时同步阻塞到出图，响应里 result_url 直接为图片 URL（status=completed）。

Input Schema

NameRequiredDescriptionDefault
modelNo可选。默认 `nanobanana-pro`。
ratioNo取值集因 model 而异，详见请求体表格。
promptNo生成 / 编辑指令
sync_modNo`true` 同步阻塞到出图（最长约 60 秒）；默认 `false` 异步返回 `workspace_id`。
image_urlsNo文生图时为空数组；编辑时非空，上限因 model 而异。
resolutionNo仅 `nanobanana-pro` / `nano-banana-2` / `nano-banana-2-fast` / `seedream-5.0-pro` / `gpt-image-2` 支持；取值集详见请求体表格。
callback_idNo自定义追踪 ID，在响应中原样回传
idempotency_keyNo幂等键（预留字段，当前服务端未对本接口做去重）

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed13 schema fields changed
    • changedInput schema / description
      Previous value: -"图片生成或编辑。`model` 为模型短名称，省略时服务端默认 `nanobanana-pro`。\n文生图：`image_urls` 为空；图片编辑：`image_urls` 非空。\n\n**模式支持**：\n- 仅文生图：`grok-2-image`\n- 仅编辑：`seedream-4.5-seq`、`seedream-5.0-lite-seq`、`grok-imagine-image`\n- 文生 + 编辑：`nanobanana-pro`、`nano-banana-2`、`nano-banana-2-fast`、`seedream-4.5`、`seedream-5.0-lite`、`gpt-image-2`\n\n**参数支持（按模型不同）**：\n- `ratio` 支持值因模型而异：\n  - `nanobanana-pro` / `nano-banana-2` / `nano-banana-2-fast`: `16:9` / `1:1` / `9:16` / `21:9` / `2:3` / `3:2`\n  - `seedream-4.5` / `seedream-4.5-seq` / `seedream-5.0-lite` / `seedream-5.0-lite-seq`: `16:9` / `1:1` / `9:16` / `2:3` / `3:2`\n  - `gpt-image-2`: `1:1` / `3:2` / `2:3` / `3:4` / `4:3` / `4:5` / `5:4` / `9:16` / `16:9` / `21:9`\n  - `grok-2-image` / `grok-imagine-image`: 不校验\n- `resolution` 仅以下模型支持，且取值集不同：\n  - `nanobanana-pro`: `1k` / `2k` / `4k` / `8k`\n  - `nano-banana-2`: `0.5k` / `1k` / `2k` / `4k`\n  - `nano-banana-2-fast`: `2k` / `4k`\n  - `gpt-image-2`: `1k` / `2k`\n- `quality` 仅 `gpt-image-2` 支持，取值：`low` / `medium` / `high`\n- `num_images` 仅 `seedream-4.5-seq` / `seedream-5.0-lite-seq` 支持，上限 15\n"New value: +"图片生成或编辑。`model` 为模型短名称，省略时服务端默认 `nanobanana-pro`。\n文生图：`image_urls` 为空；图片编辑：`image_urls` 非空。每次固定生成 1 张。\n\n**模式支持**：\n- 仅文生图：`grok-2-image`\n- 仅编辑：`grok-imagine-image`\n- 文生 + 编辑：`nanobanana-pro`、`nano-banana-2`、`nano-banana-2-fast`、`seedream-5.0-pro`、`seedream-5.0-lite`、`gpt-image-2`\n\n**参数支持（按模型不同，服务端校验为准）**：\n\n| model | ratio | resolution | 编辑输入图上限 |\n|---|---|---|---|\n| `nanobanana-pro` | 1:1 / 3:2 / 2:3 / 3:4 / 4:3 / 4:5 / 5:4 / 5:7 / 7:5 / 9:16 / 16:9 / 21:9 / 1:4 / 4:1 | 1k / 2k / 4k / 8k | 14 |\n| `nano-banana-2` | 同上 + 1:8 / 8:1 | 0.5k / 1k / 2k / 4k | 14 |\n| `nano-banana-2-fast` | 同上 + 1:8 / 8:1 | 2k / 4k | 14 |\n| `seedream-5.0-pro` | 1:1 / 1:2 / 2:1 / 1:3 / 3:1 / 2:3 / 3:2 / 3:4 / 4:3 / 4:5 / 5:4 / 9:16 / 16:9 / 9:21 / 21:9 | 2k（默认）/ 1k | 10 |\n| `seedream-5.0-lite` | 1:1 / 1:2 / 2:1 / 1:3 / 3:1 / 1:4 / 4:1 / 1:8 / 8:1 / 2:3 / 3:2 / 3:4 / 4:3 / 4:5 / 5:4 / 5:7 / 7:5 / 9:16 / 16:9 / 9:21 / 21:9 | — | 10 |\n| `gpt-image-2` | 1:1 / 1:2 / 2:1 / 1:3 / 3:1 / 2:3 / 3:2 / 3:4 / 4:3 / 4:5 / 5:4 / 9:16 / 16:9 / 9:21 / 21:9 | 1k / 2k / 4k | 10 |\n| `grok-2-image` | 不校验 | — | 不支持编辑 |\n| `grok-imagine-image` | 1:1 / 3:2 / 2:3 / 3:4 / 4:3 / 4:5 / 5:4 / 9:16 / 16:9 / 21:9 | — | 1 |\n\n- `ratio` 留空由模型按 prompt 决定；prompt 内写明的比例（如 `16:9`、「竖屏」）优先于 `ratio` 参数。\n- `resolution` 不支持的模型必须留空；`4k` 及以上需订阅身份。\n- `image_urls` 只收 http(s) URL，不收 data URI；单张像素 ≤ 3600 万。\n"
    • addedInput schema / properties / callback_id
      Added value: +{
+  "description": "自定义追踪 ID，在响应中原样回传",
+  "type": "string"
+}
    • removedInput schema / properties / concurrency
      Removed value: -{
-  "description": "并发数，控制一次性输出的结果数量",
-  "example": 1,
-  "format": "int32",
-  "type": "integer"
-}
    • addedInput schema / properties / idempotency_key
      Added value: +{
+  "description": "幂等键（预留字段，当前服务端未对本接口做去重）",
+  "type": "string"
+}
    • changedInput schema / properties / image_urls / description
      Previous value: -"文生图时为空数组；编辑时非空。\n"New value: +"文生图时为空数组；编辑时非空，上限因 model 而异。\n"
    • changedInput schema / properties / model / enum
      Previous value: -[
-  "nanobanana-pro",
-  "nano-banana-2",
-  "nano-banana-2-fast",
-  "seedream-4.5",
-  "seedream-4.5-seq",
-  "seedream-5.0-lite",
-  "seedream-5.0-lite-seq",
-  "gpt-image-2",
-  "grok-2-image",
-  "grok-imagine-image"
-]New value: +[
+  "nanobanana-pro",
+  "nano-banana-2",
+  "nano-banana-2-fast",
+  "seedream-5.0-pro",
+  "seedream-5.0-lite",
+  "gpt-image-2",
+  "grok-2-image",
+  "grok-imagine-image"
+]
    • removedInput schema / properties / num_images
      Removed value: -{
-  "description": "仅 `seedream-4.5-seq` / `seedream-5.0-lite-seq` 支持，单次输出图片数量。",
-  "format": "int32",
-  "maximum": 15,
-  "minimum": 1,
-  "type": "integer"
-}
    • changedInput schema / properties / prompt / description
      Previous value: -"文生图提示词"New value: +"生成 / 编辑指令"
    • removedInput schema / properties / quality
      Removed value: -{
-  "description": "仅 `gpt-image-2` 支持。",
-  "enum": [
-    "low",
-    "medium",
-    "high"
-  ],
-  "type": "string"
-}
    • changedInput schema / properties / ratio / description
      Previous value: -"取值集因 model 而异，详见请求体描述。"New value: +"取值集因 model 而异，详见请求体表格。"
    • changedInput schema / properties / ratio / enum
      Previous value: -[
-  "1:1",
-  "3:2",
-  "2:3",
-  "3:4",
-  "4:3",
-  "4:5",
-  "5:4",
-  "16:9",
-  "9:16",
-  "21:9"
-]New value: +[
+  "1:1",
+  "1:2",
+  "2:1",
+  "1:3",
+  "3:1",
+  "1:4",
+  "4:1",
+  "1:8",
+  "8:1",
+  "2:3",
+  "3:2",
+  "3:4",
+  "4:3",
+  "4:5",
+  "5:4",
+  "5:7",
+  "7:5",
+  "9:16",
+  "16:9",
+  "9:21",
+  "21:9"
+]
    • changedInput schema / properties / resolution / description
      Previous value: -"仅 `nanobanana-pro` / `nano-banana-2` / `nano-banana-2-fast` / `gpt-image-2` 支持；取值集详见请求体描述。\n"New value: +"仅 `nanobanana-pro` / `nano-banana-2` / `nano-banana-2-fast` / `seedream-5.0-pro` / `gpt-image-2` 支持；取值集详见请求体表格。\n"
    • addedInput schema / properties / sync_mod / description
      Added value: +"`true` 同步阻塞到出图（最长约 60 秒）；默认 `false` 异步返回 `workspace_id`。"
  2. First observed

TDQS

A3.7/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations declare readOnlyHint=false, openWorldHint=true and idempotentHint=false, but say nothing about the call lifecycle. The description adds genuinely useful behavior: the default is asynchronous with a workspace_id/status=pending handoff, polling with work/status is required to obtain result_url, and sync_mod=true blocks for roughly 60 seconds. That polling contract is exactly the kind of trait an agent needs and cannot infer from annotations.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Three short lines, front-loaded with purpose, then the fixed one-image scope, then the mode behaviour — no wasted sentences. It is dense with inline code formatting but every clause earns its place.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

With 8 optional parameters, no output schema, and a complex async contract, the description compensates well by spelling out the return shape (workspace_id, status, result_type, result_url) and the polling requirement. The model/ratio/resolution support matrices live in the schema, so omitting them from the description is acceptable, though the absence of any return-value detail for the sync path beyond result_url is a minor gap.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100% and the schema itself carries extensive per-model tables for ratio, resolution, and image_urls limits, so the baseline is 3. The description adds only the async/sync interpretation of sync_mod and the meaning of the returned result_url; it adds no parameter semantics beyond what the schema already documents.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource up front ('图片生成 / 编辑') and pins down the scope with '每次调用固定生成 1 张', which is a concrete distinguishing constraint. It does not, however, differentiate itself from sibling editing tools like image_cutout, image_enhance, or image_poster, so an agent can't tell from the description alone when this general tool beats a specialized one.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description gives real conditional guidance on mode selection: default async returns workspace_id and must be polled via work/status, while sync_mod=true blocks until the image is ready. It offers no guidance on tool selection versus the many sibling image tools, so usage is only implied at the tool level.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All AdsTurbo toolsInspect
Try in Browser
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources