# Gemini Models API — Model Query

Source: https://ofox.io/docs/api/gemini/models

Query available models on the OfoxAI platform through the Gemini native protocol. The response format follows the Google GenAI API specification (`name`, `displayName`, `inputTokenLimit`, and other camelCase fields).

The Models API is a public endpoint that does not require an API Key. This endpoint returns all models accessible via the Gemini protocol, not limited to Google models.

## List All Models

```http
GET https://api.ofox.io/gemini/v1beta/models
```

### Request Example

```bash filename="Terminal"
curl https://api.ofox.io/gemini/v1beta/models
```

### Response Format

```json
{
  "models": [
    {
      "name": "models/google/gemini-3.8-flash",
      "version": "001",
      "displayName": "Google: Gemini 3.8 Flash",
      "description": "Gemini 3.8 Flash is Google's most intelligent Flash model, with significant gains over 3.7 Flash across software engineering, agentic tasks, and multi-step reasoning. All input modalities (text/image/video/audio) unified pricing; batch tier at 50% discount. Thinking model with configurable levels. 1M context. Released September 2, 2026.",
      "inputTokenLimit": 1000000,
      "outputTokenLimit": 65536,
      "supportedGenerationMethods": [
        "generateContent",
        "countTokens"
      ],
      "ownedBy": "google",
      "canonicalSlug": "gemini-3.8-flash",
      "contextLength": 1000000,
      "architecture": {
        "modality": "text+image+audio+file->text",
        "input_modalities": [
          "text",
          "image",
          "audio",
          "file"
        ],
        "output_modalities": [
          "text"
        ],
        "tokenizer": "gemini",
        "instruct_type": null
      },
      "pricing": {
        "prompt": "0.0000015",
        "completion": "0.0000075",
        "audio": "0.0000015",
        "input_cache_read": "0.00000015",
        "input_cache_write": "0.000000083",
        "input_cache_write_1h": "0.000001",
        "input_cached_audio": "0.00000015",
        "web_search": "0.014"
      },
      "topProvider": {
        "context_length": 1000000,
        "max_completion_tokens": 65536,
        "is_moderated": false
      },
      "supportedParameters": [
        "temperature",
        "top_p",
        "max_tokens",
        "stop",
        "tools",
        "tool_choice",
        "response_format",
        "reasoning"
      ],
      "expirationDate": null,
      "isDeprecated": false,
      "aliases": [
        "gemini-3.8-flash"
      ]
    }
  ]
}
```

### Field Differences from OpenAI Protocol

| Field | OpenAI Protocol | Gemini Protocol |
|-------|----------------|-----------------|
| List key | `data` | `models` |
| Model identifier | `id` | `name` (with `models/` prefix) |
| Name | `name` | `displayName` |
| Context length | `context_length` | `inputTokenLimit` + `contextLength` |
| Max output | `top_provider.max_completion_tokens` | `outputTokenLimit` |
| Naming style | snake_case | camelCase |

## Filter Models by Provider

```http
GET https://api.ofox.io/gemini/v1beta/models/{provider}
```

### Path Parameters

| Parameter | Type | Description |
|-----------|------|-------------|
| `provider` | string | Provider identifier, e.g. `google`, `anthropic`, `openai` |

### Request Example

```bash filename="Terminal"
# List only Google models
curl https://api.ofox.io/gemini/v1beta/models/google
```

### Response Format

The response format is the same as listing all models, with the `models` array containing only models from the specified provider.

## Get Model Details

```http
GET https://api.ofox.io/gemini/v1beta/models/{provider}/{model_id}
```

### Path Parameters

| Parameter | Type | Description |
|-----------|------|-------------|
| `provider` | string | Provider identifier, e.g. `google` |
| `model_id` | string | Model name, e.g. `gemini-3.8-flash` |

### Request Example

```bash filename="Terminal"
curl https://api.ofox.io/gemini/v1beta/models/google/gemini-3.8-flash
```

### Response Format

Returns a single Model object with the same structure as objects in the list endpoint.