3.26.2 (#325)

hannesrudolph · web-flow · commit e126e0b48983 · 2025-08-28T13:00:31.000-06:00
* 3.26.2

* feat: restore 'image-generation' feature in sidebars configuration
diff --git a/docs/features/experimental/experimental-features.md b/docs/features/experimental/experimental-features.md
@@ -35,6 +35,7 @@ The following experimental features are currently available:
 - [Concurrent File Edits](/features/experimental/concurrent-file-edits) - Edit multiple files in a single operation
 - [Power Steering](/features/experimental/power-steering) - Enhanced consistency in AI responses
 - [Background Editing](/features/experimental/background-editing) - Work uninterrupted while Roo edits files in the background
+- [Image Generation](/features/image-generation) - Generate images from text prompts and save them to your workspace
 
 ---
 
diff --git a/docs/features/image-generation.md b/docs/features/image-generation.md
@@ -0,0 +1,106 @@
+---
+description: Generate images from text prompts directly in Roo Code using OpenRouter API. Save AI-generated images to your workspace with preview support.
+keywords:
+  - image generation
+  - text to image
+  - OpenRouter
+  - AI images
+  - experimental feature
+  - image creation
+  - prompt to image
+image: /img/social-share.jpg
+---
+
+# Image Generation
+
+Generate images from plain text prompts, save them to your project, and preview them in the chat. This experimental feature requires an OpenRouter API key.
+
+:::warning Experimental Feature
+Image Generation is an experimental feature that requires enabling in settings and configuring an OpenRouter API key.
+:::
+
+---
+
+## Key Features
+
+- Create images directly from chat using natural language
+- Saves to your workspace at a path you choose; file extension is auto-added if missing
+- Shows a preview of the generated image in the conversation
+- Choose an image generation model (via OpenRouter)
+- Simple on/off toggle under Experimental settings
+
+---
+
+## Use Case
+
+**Before:** You had to copy prompts to an external site, download the result, then move it into your workspace.
+
+**With this feature:** Ask Roo to generate an image, approve, pick a save location, and continue editing with the image already in your project.
+
+---
+
+## How It Works
+
+When enabled, Roo sends your prompt to an image-capable model through OpenRouter. The image returned by OpenRouter is saved to the path you specify inside your current workspace. Roo shows a preview in the chat and the file appears in your file explorer.
+
+---
+
+## Requirements
+
+- OpenRouter account and API key
+- Internet access
+- An open, writable workspace folder
+
+---
+
+## Configuration
+
+### 1. Enable Image Generation (Experimental)
+- **Purpose:** Turns the feature on so Roo can create images on request
+- **Default:** Off
+- **Location:** Settings > Experimental
+
+### 2. OpenRouter API Key
+- **Purpose:** Authorizes image generation requests
+- **Default:** Empty (required)
+- **Get your key:** [https://openrouter.ai/keys](https://openrouter.ai/keys)
+
+### 3. Image Generation Model
+- **Purpose:** Selects which model to use for generation
+- **Default:** Gemini 2.5 Flash Image Preview (or the free preview variant)
+
+---
+
+## Using Image Generation
+
+1. In chat, ask Roo to generate an image and describe what you want (subject, style, lighting, composition).
+2. Confirm the action when prompted. Roo may ask you to choose a save path (for example: `images/sunset.png`).
+3. Roo generates the image and saves it. If you don't include an extension, one is added automatically (png or jpg).
+4. See the image preview in the chat and locate the file in your workspace.
+
+---
+
+## Tips for Better Results
+
+### Be Specific
+Include these elements in your prompts:
+- **Style:** artistic medium, art movement, or specific artist style
+- **Mood:** emotional tone, atmosphere
+- **Color palette:** specific colors or color schemes
+- **Camera/lighting:** angle, perspective, lighting conditions
+- **Aspect ratio:** dimensions or orientation
+
+---
+
+## Limitations
+
+- Experimental feature; availability and model list are limited
+- One image is produced per request
+- Output formats supported: PNG or JPG
+- Usage may be subject to your OpenRouter plan limits and costs
+
+---
+
+## Status
+
+This feature is experimental and may change or be removed in future versions. Provide feedback through [GitHub Issues](https://github.com/RooCodeInc/Roo-Code/issues).
diff --git a/docs/update-notes/index.md b/docs/update-notes/index.md
@@ -19,6 +19,7 @@ image: /img/social-share.jpg
 
 ### Version 3.26
 
+*   [3.26.2](/update-notes/v3.26.2) (2025-08-28)
 *   [3.26.1](/update-notes/v3.26.1) (2025-08-27)
 *   [3.26.0](/update-notes/v3.26.0) (2025-08-26)
 
diff --git a/docs/update-notes/v3.26.2.mdx b/docs/update-notes/v3.26.2.mdx
@@ -0,0 +1,69 @@
+---
+description: Adds built-in image generation (OpenRouter), a free Gemini preview option, GPT-5 usage without BYOK rate limit blockers, plus QOL improvements and fixes.
+keywords:
+  - roo code 3.26.2
+  - release notes
+  - image generation
+  - openrouter
+  - gemini 2.5 flash image preview
+  - free image generation
+  - prompt to image
+  - image viewer
+  - approvals flow
+  - gpt-5
+  - rate limits
+  - byok
+  - openai provider
+  - model picker
+  - input modalities
+  - output modalities
+  - provider updates
+  - qol improvements
+  - bug fixes
+  - release automation
+  - taskspawned event
+image: /img/social-share.jpg
+---
+
+# Roo Code 3.26.2 Release Notes (2025-08-28)
+
+This release adds integrated image generation, a free Gemini image preview option, improved GPT-5 availability without BYOK rate limit blockers, and multiple quality-of-life improvements and fixes.
+
+## Image Generation (OpenRouter) — Free option: Gemini 2.5 Flash Image Preview
+
+Generate images from natural‑language prompts directly inside Roo Code using OpenRouter’s image generation models. Configure your OpenRouter API key, pick a supported model, and preview results in the built‑in Image Viewer. See [Image Generation](/features/image-generation) and [OpenRouter Provider](/providers/openrouter) for setup and model selection.
+
+- **Free option available: Gemini 2.5 Flash Image Preview** — try image generation without paid credits for faster onboarding and quick experiments
+- Prompt‑to‑image workflow inside the editor with approvals flow (supports auto‑approval when write permissions are granted)
+- Image Viewer with zoom, copy, and save for quick reuse in docs and prototypes
+
+PRs: [#7474](https://github.com/RooCodeInc/Roo-Code/pull/7474), [#7492](https://github.com/RooCodeInc/Roo-Code/pull/7492), [#7493](https://github.com/RooCodeInc/Roo-Code/pull/7493)
+
+
+## GPT-5 usage without BYOK rate limit blockers
+
+If you’re being rate limited with GPT‑5, you can now use GPT‑5 models without bringing your own key. This improves availability and reduces interruptions during development. See [OpenAI Provider](/providers/openai) for model support and guidance.
+
+
+## QOL Improvements
+
+- Improved padding and click targets in the image model picker for easier selection and fewer misclicks. (#[7494](https://github.com/RooCodeInc/Roo-Code/pull/7494))
+- Generic default filename for saved images (e.g., `img_<timestamp>`) instead of `mermaid_diagram_<timestamp>`. (#[7479](https://github.com/RooCodeInc/Roo-Code/pull/7479))
+
+## Bug Fixes
+
+- ImageGenerationSettings no longer shows a dirty state on first open; the save button only enables after an actual change. (#[7495](https://github.com/RooCodeInc/Roo-Code/pull/7495))
+- GPT‑5 reliability improvements:
+  - Manual condense preserves conversation continuity by correctly handling `previous_response_id` on the next request
+  - Image inputs work reliably with structured text+image payloads
+  - Temperature control is shown only for models that support it
+  - Fewer GPT‑5–specific errors with updated provider definitions and SDK (thanks nlbuescher!)
+
+  (#[7067](https://github.com/RooCodeInc/Roo-Code/pull/7067))
+
+
+## Misc Improvements
+
+- Release automation: version bumps, changelog updates, and auto-publishing on merge for a faster, more reliable release process. (#[7490](https://github.com/RooCodeInc/Roo-Code/pull/7490))
+- New TaskSpawned developer event so integrations can detect when a subtask is created and capture its ID for chaining or monitoring. (#[7465](https://github.com/RooCodeInc/Roo-Code/pull/7465))
+- Roo Code Cloud SDK bumped to 0.25.0. (#[7475](https://github.com/RooCodeInc/Roo-Code/pull/7475))
diff --git a/docs/update-notes/v3.26.mdx b/docs/update-notes/v3.26.mdx
@@ -67,6 +67,16 @@ We've added Vercel AI Gateway as a complete provider integration (thanks joshual
 
 > **📚 Documentation**: See [Vercel AI Gateway](/providers/vercel-ai-gateway) for detailed setup instructions.
 
+### Image Generation (OpenRouter) — Free option: Gemini 2.5 Flash Image Preview
+
+Generate images from natural‑language prompts directly inside Roo Code using OpenRouter’s image generation models. Configure your OpenRouter API key, pick a supported model, and preview results in the built‑in Image Viewer. See [Image Generation](/features/image-generation) and [OpenRouter Provider](/providers/openrouter) for setup and model selection.
+
+- **Free option available: Gemini 2.5 Flash Image Preview** — try image generation without paid credits for faster onboarding and quick experiments
+- Prompt‑to‑image workflow inside the editor with approvals flow (supports auto‑approval when write permissions are granted)
+- Image Viewer with zoom, copy, and save for quick reuse in docs and prototypes
+
+PRs: [#7474](https://github.com/RooCodeInc/Roo-Code/pull/7474), [#7492](https://github.com/RooCodeInc/Roo-Code/pull/7492), [#7493](https://github.com/RooCodeInc/Roo-Code/pull/7493)
+
 ### Provider Updates
 
 * **DeepSeek V3.1 on Fireworks**: Added support for DeepSeek V3.1 model in the Fireworks AI provider (thanks dmarkey!) ([#7375](https://github.com/RooCodeInc/Roo-Code/pull/7375))
@@ -82,6 +92,10 @@ We've added Vercel AI Gateway as a complete provider integration (thanks joshual
 * **Model ID Visibility**: API configuration dropdown now shows model IDs alongside profile names for easier identification ([#7423](https://github.com/RooCodeInc/Roo-Code/pull/7423))
 * **Chat UI Cleanup**: Improved consistency in chat input controls and fixed tooltip behavior ([#7436](https://github.com/RooCodeInc/Roo-Code/pull/7436))
 * **Clearer Task Headers**: Removed duplicate cache display in task headers to eliminate confusion ([#7443](https://github.com/RooCodeInc/Roo-Code/pull/7443))
+* Improved padding and click targets in the image model picker for easier selection and fewer misclicks. (#[7494](https://github.com/RooCodeInc/Roo-Code/pull/7494))
+* Generic default filename for saved images (e.g., `img_<timestamp>`) instead of `mermaid_diagram_<timestamp>`. (#[7479](https://github.com/RooCodeInc/Roo-Code/pull/7479))
+* Improved padding and click targets in the image model picker for easier selection and fewer misclicks. (#[7494](https://github.com/RooCodeInc/Roo-Code/pull/7494))
+* Generic default filename for saved images (e.g., `img_<timestamp>`) instead of `mermaid_diagram_<timestamp>`. (#[7479](https://github.com/RooCodeInc/Roo-Code/pull/7479))
 
 ### Bug Fixes
 
@@ -92,9 +106,31 @@ We've added Vercel AI Gateway as a complete provider integration (thanks joshual
 * **Mistral Thinking Content**: Fixed validation errors when using Mistral models that send thinking content (thanks Biotrioo!) ([#7106](https://github.com/RooCodeInc/Roo-Code/pull/7106))
 * **Requesty Model Listing**: Fixed model listing for Requesty provider when using custom base URLs (thanks dtrugman!) ([#7378](https://github.com/RooCodeInc/Roo-Code/pull/7378))
 * **Todo List Setting**: Fixed newTaskRequireTodos setting to properly enforce todo list requirements ([#7363](https://github.com/RooCodeInc/Roo-Code/pull/7363))
+* ImageGenerationSettings no longer shows a dirty state on first open; the save button only enables after an actual change. (#[7495](https://github.com/RooCodeInc/Roo-Code/pull/7495))
+* GPT‑5 reliability improvements:
+  - Manual condense preserves conversation continuity by correctly handling `previous_response_id` on the next request
+  - Image inputs work reliably with structured text+image payloads
+  - Temperature control is shown only for models that support it
+  - Fewer GPT‑5–specific errors with updated provider definitions and SDK (thanks nlbuescher!)
+
+  (#[7067](https://github.com/RooCodeInc/Roo-Code/pull/7067))
+* ImageGenerationSettings no longer shows a dirty state on first open; the save button only enables after an actual change. (#[7495](https://github.com/RooCodeInc/Roo-Code/pull/7495))
+* GPT‑5 reliability improvements:
+  - Manual condense preserves conversation continuity by correctly handling `previous_response_id` on the next request
+  - Image inputs work reliably with structured text+image payloads
+  - Temperature control is shown only for models that support it
+  - Fewer GPT‑5–specific errors with updated provider definitions and SDK (thanks nlbuescher!)
+
+  (#[7067](https://github.com/RooCodeInc/Roo-Code/pull/7067))
 
 ### Misc Improvements
 
 * **Issue Fixer Mode**: Added missing todos parameter in new_task tool usage ([#7391](https://github.com/RooCodeInc/Roo-Code/pull/7391))
 * **Privacy Policy Update**: Updated privacy policy to clarify proxy mode data handling (thanks jdilla1277!) ([#7255](https://github.com/RooCodeInc/Roo-Code/pull/7255))
-* **Dependencies**: Updated drizzle-kit to v0.31.4 ([#5453](https://github.com/RooCodeInc/Roo-Code/pull/5453))
+* **Dependencies**: Updated drizzle-kit to v0.31.4 ([#5453](https://github.com/RooCodeInc/Roo-Code/pull/5453))
+* Release automation: version bumps, changelog updates, and auto-publishing on merge for a faster, more reliable release process. (#[7490](https://github.com/RooCodeInc/Roo-Code/pull/7490))
+* New TaskSpawned developer event so integrations can detect when a subtask is created and capture its ID for chaining or monitoring. (#[7465](https://github.com/RooCodeInc/Roo-Code/pull/7465))
+* Roo Code Cloud SDK bumped to 0.25.0. (#[7475](https://github.com/RooCodeInc/Roo-Code/pull/7475))
+* Release automation: version bumps, changelog updates, and auto-publishing on merge for a faster, more reliable release process. (#[7490](https://github.com/RooCodeInc/Roo-Code/pull/7490))
+* New TaskSpawned developer event so integrations can detect when a subtask is created and capture its ID for chaining or monitoring. (#[7465](https://github.com/RooCodeInc/Roo-Code/pull/7465))
+* Roo Code Cloud SDK bumped to 0.25.0. (#[7475](https://github.com/RooCodeInc/Roo-Code/pull/7475))
diff --git a/sidebars.ts b/sidebars.ts
@@ -106,6 +106,7 @@ const sidebars: SidebarsConfig = {
             'features/experimental/concurrent-file-edits',
             'features/experimental/power-steering',
             'features/experimental/background-editing',
+            'features/image-generation',
           ],
         },
         'features/more-features',
@@ -216,6 +217,7 @@ const sidebars: SidebarsConfig = {
           label: '3.26',
           items: [
             { type: 'doc', id: 'update-notes/v3.26', label: '3.26 Combined' },
+            { type: 'doc', id: 'update-notes/v3.26.2', label: '3.26.2' },
             { type: 'doc', id: 'update-notes/v3.26.1', label: '3.26.1' },
             { type: 'doc', id: 'update-notes/v3.26.0', label: '3.26.0' },
           ],
