[BOT ISSUE] Google: add missing gemini-omni-flash-preview model

Start in model_list.json near the existing Gemini image and video model entries listed in the issue, then compare their catalog fields with the supplied model specification. Add gemini-omni-flash-preview using the verified Google pricing and model documentation, preserving the noted uncertainty around token limits and video pricing. Done means the model is present with the documented metadata and no duplicate entry remains.

Gap description

The Google Gemini model gemini-omni-flash-preview is listed on the Google AI pricing page and models page, but is missing from the catalog. The catalog already includes similar multimodal generation models like gemini-3.1-flash-image, gemini-3.1-flash-lite-image, gemini-3-pro-image, and gemini-2.5-flash-image.

Missing model ID

Google describes this model as offering "Fast, conversational video generation and editing" — it accepts text and image input and can output both text and video. It is listed as a "New Preview" under the Gemini 3 section.

Suggested catalog entry

"gemini-omni-flash-preview": {
  "format": "google",
  "flavor": "chat",
  "multimodal": true,
  "input_cost_per_mil_tokens": 1.5,
  "output_cost_per_mil_tokens": 9,
  "displayName": "Gemini Omni Flash Preview",
  "experimental": true,
  "available_providers": ["google"]
}

Notes on specific fields

• output_cost_per_mil_tokens: Set to $9.00 (text output rate). Video output at $17.50/M tokens is a separate modality and may need a different field or documentation note.
• max_input_tokens / max_output_tokens: Not published on the pricing or models page. Should be verified from the model detail page or API docs.
• context window: Not available in sources checked — omitted until verified.

Sources

• Google AI Developer Pricing — lists gemini-omni-flash-preview with pricing: input $1.50, text output $9.00, video output $17.50
• Google AI Model Cards — lists the model as "New Preview" under Gemini 3, described as "Fast, conversational video generation and editing"

Verification checklist

• Cross-source: Model confirmed on (1) Google AI pricing page AND (2) Google AI models docs page
• Already-fixed check: Grepped model_list.json for gemini-omni-flash — no matches
• ID format check: Follows the standard Gemini naming convention (gemini-{version}-{variant})
• Duplicate issue check: Searched open issues for "gemini-omni-flash" — no results
• Not excluded: Not embedding, not computer-vision-only (accepts text, outputs text + video), no retirement date

Catalog precedent

The catalog includes similar Gemini image/video generation models:

• gemini-3.1-flash-image (line 10940)
• gemini-3.1-flash-lite-image (line 11003)
• gemini-3-pro-image (line 11112)
• gemini-2.5-flash-image (line 11235)

{
  "kind": "missing_model",
  "provider": "google",
  "models": ["gemini-omni-flash-preview"],
  "model_specs": {
    "gemini-omni-flash-preview": {
      "format": "google",
      "flavor": "chat",
      "multimodal": true,
      "input_cost_per_mil_tokens": 1.5,
      "output_cost_per_mil_tokens": 9,
      "displayName": "Gemini Omni Flash Preview",
      "experimental": true,
      "available_providers": ["google"]
    }
  },
  "source_urls": [
    "https://ai.google.dev/pricing",
    "https://ai.google.dev/gemini-api/docs/models"
  ]
}