Model API announcements

Archived pricing and model announcements, plus setup guides for OpenClaw, Claude Code and Codex. Check your account dashboard for current prices and model availability.

Update centre

The upstream cost of Gemini image generation increased after changes to the platform’s risk controls.

Please check your usage Review the token counts in your request logs and update your budget. Contact support if you have questions.

Google’s gemini-3.1-flash-image-preview image generation model is available on the platform.

gpt-image-2 is now available.

Release highlights The launch announcement listed 2K and 4K output, unrestricted call volume and introductory pricing.

Google introduced nano-banana-2 as a faster second-generation image model with stronger consistency across outputs.

🖼️
High-resolution output Up to 4K (4096 × 4096), matching the Pro version.
🎭
Reference image consistency Up to 14 references, including 10 objects and 4 characters, help preserve style and character identity.
📐
Extended aspect ratios 1:4, 4:1, 1:8 and 8:1 for long layouts, posters and banners.
🔤
Text rendering Improved text generation for infographics and marketing layouts.
🔍
Search grounding Google Search and Image Search grounding.
🧠
Reasoning A built-in reasoning step helps the model work through complex prompts before generation.

Supported API formats

Gemini API OpenAI Chat OpenAI DALL·E

Find the configuration file in the installation directory.

📁 /openclaw/openclaw/data/conf/openclaw.json

Add the following settings.

openclaw.json
{
  "env": {
    "mola-api key": "YOUR_API_KEY"  // Replace with your API key
  },
  "agents": {
    "defaults": {
      "model": {
        "primary": "mola/gpt-4o"
      }
    }
  },
  "models": {
    "providers": {
      "mola": {
        "baseUrl": "https://api4.mygptlife.com/v1",
        "apiKey": "${MOLA_API_KEY}",
        "api": "openai-completions",
        "models": [{
          "id": "gpt-4o",
          "name": "gpt-4o"
        }]
      }
    }
  }
}