@@ -1,5 +1,5 @@
11---
2-summary: "Generate and edit images using configured providers (OpenAI, Google Gemini, fal, MiniMax, ComfyUI, Vydra, xAI)"
2+summary: "Generate and edit images using configured providers (OpenAI, OpenAI Codex OAuth, Google Gemini, fal, MiniMax, ComfyUI, Vydra, xAI)"
33read_when:
44 - Generating images via the agent
55 - Configuring image generation providers and models
@@ -10,12 +10,12 @@ title: "Image generation"
1010The `image_generate` tool lets the agent create and edit images using your configured providers. Generated images are delivered automatically as media attachments in the agent's reply.
11111212<Note>
13-The tool only appears when at least one image generation provider is available. If you don't see `image_generate` in your agent's tools, configure `agents.defaults.imageGenerationModel` or set up a provider API key.
13+The tool only appears when at least one image generation provider is available. If you don't see `image_generate` in your agent's tools, configure `agents.defaults.imageGenerationModel`, set up a provider API key, or sign in with OpenAI Codex OAuth.
1414</Note>
15151616## Quick start
171718-1. Set an API key for at least one provider (for example `OPENAI_API_KEY` or `GEMINI_API_KEY`).
18+1. Set an API key for at least one provider (for example `OPENAI_API_KEY` or `GEMINI_API_KEY`) or sign in with OpenAI Codex OAuth.
19192. Optionally set your preferred model:
20202121```json5
@@ -30,6 +30,20 @@ The tool only appears when at least one image generation provider is available.
3030}
3131```
323233+Use Codex OAuth instead of an OpenAI API key:
34+35+```json5
36+{
37+ agents: {
38+ defaults: {
39+ imageGenerationModel: {
40+ primary: "openai-codex/gpt-image-2",
41+ },
42+ },
43+ },
44+}
45+```
46+33473. Ask the agent: _"Generate an image of a friendly lobster mascot."_
34483549The agent calls `image_generate` automatically. No tool allow-listing needed — it's enabled by default when a provider is available.
@@ -59,7 +73,7 @@ Use `action: "list"` to inspect available providers and models at runtime:
5973| ------------- | -------- | ------------------------------------------------------------------------------------- |
6074| `prompt` | string | Image generation prompt (required for `action: "generate"`) |
6175| `action` | string | `"generate"` (default) or `"list"` to inspect providers |
62-| `model` | string | Provider/model override, e.g. `openai/gpt-image-2` |
76+| `model` | string | Provider/model override, e.g. `openai/gpt-image-2` or `openai-codex/gpt-image-2` |
6377| `image` | string | Single reference image path or URL for edit mode |
6478| `images` | string[] | Multiple reference images for edit mode (up to 5) |
6579| `size` | string | Size hint: `1024x1024`, `1536x1024`, `1024x1536`, `2048x2048`, `3840x2160` |
@@ -125,9 +139,11 @@ OpenAI, Google, and xAI support up to 5 reference images via the `images` parame
125139126140### OpenAI `gpt-image-2`
127141128-OpenAI image generation defaults to `openai/gpt-image-2`. The older
129-`openai/gpt-image-1` model can still be selected explicitly, but new OpenAI
130-image-generation and image-editing requests should use `gpt-image-2`.
142+OpenAI image generation defaults to `openai/gpt-image-2` with `OPENAI_API_KEY`.
143+Use `openai-codex/gpt-image-2` to generate or edit images with the same Codex
144+OAuth sign-in used by `openai-codex` chat models. The older `openai/gpt-image-1`
145+model can still be selected explicitly, but new OpenAI image-generation and
146+image-editing requests should use `gpt-image-2`.
131147132148`gpt-image-2` supports both text-to-image generation and reference-image
133149editing through the same `image_generate` tool. OpenClaw forwards `prompt`,
@@ -153,6 +169,18 @@ Edit one local reference image:
153169/tool image_generate action=generate model=openai/gpt-image-2 prompt="Keep the subject, replace the background with a bright studio setup" image=/path/to/reference.png size=1024x1536
154170```
155171172+Generate with Codex OAuth:
173+174+```
175+/tool image_generate action=generate model=openai-codex/gpt-image-2 prompt="A clean editorial poster for OpenClaw image generation" size=3840x2160 count=1
176+```
177+178+Edit one local reference image with Codex OAuth:
179+180+```
181+/tool image_generate action=generate model=openai-codex/gpt-image-2 prompt="Keep the subject, replace the background with a bright studio setup" image=/path/to/reference.png size=1024x1536
182+```
183+156184Edit with multiple references:
157185158186```
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。