Skip to content
/
Model object
Last updated

Model object

🔨 In Development — This section is still being developed and may change.
Represents an AI model available in the Freddy platform with its capabilities, pricing, configuration details, and performance characteristics.

Properties

Core Identity

id string required

The unique identifier for the model (e.g., mdl_09adacc3e5d3).

key string required

The API key used to reference this model in requests (e.g., gpt-4o, claude-3-5-sonnet-20241022).

name string required

The human-readable display name of the model (e.g., "GPT-4o", "Claude 3.5 Sonnet").

description string optional

A brief description of the model's strengths and primary use cases.

Provider & Status

provider string required

The AI provider for this model. Like these: aitronos, openai, anthropic, google (not limited to these).

is_visible_in_ui boolean required

Whether this model is visible in the chat UI or Hub for end users.

is_deprecated boolean required

Whether this model is deprecated and scheduled for removal.

deprecation_date string optional

ISO 8601 date when the model will be removed from service. Only present if is_deprecated is true.

availability_status string optional

Current availability status. One of beta, general_availability, limited, deprecated.

Capabilities

capabilities array required

List of features supported by this model.

Show possible values
  • text_generation - Generate text responses
  • conversation - Multi-turn conversations
  • streaming - Real-time response streaming
  • tool_calling - Execute functions and tools
  • vision - Process and analyze images
  • audio - Process audio inputs
  • reasoning - Advanced reasoning capabilities
  • code_interpreter - Execute code
  • web_search - Search the internet
  • file_search - Search uploaded files
  • image_generation - Generate images
  • structured_output - JSON schema output

Context & Synapse Limits

context_window integer required

Maximum number of tokens the model can process in a single request (input context window).

Pricing

pricing object required

Pricing information for this model.

Show properties

input_synapses_cost number required

Cost per 1M input synapses (processing user input and context).

output_synapses_cost number required

Cost per 1M output synapses (generating model responses).

Dates & Versioning

release_date string optional

ISO 8601 date when the model was officially released.

model_version string optional

Version identifier for the model (e.g., 2024-11-20).

training_data_cutoff string optional

ISO 8601 date indicating the knowledge cutoff for the model's training data.

added_to_aitronos_date string optional

ISO 8601 date when this model was added to the Aitronos platform.

Recommendations & Use Cases

recommended_for array optional

List of use cases this model is recommended for.

Show possible values
  • general_purpose - General-purpose tasks
  • coding - Software development and code generation
  • analysis - Data analysis and reasoning
  • creative_writing - Creative content generation
  • multimodal - Image and text processing
  • vision - Image understanding and analysis
  • audio - Audio processing
  • tool_calling - Function calling and tool use
  • reasoning - Complex reasoning tasks
  • fast_response - Quick responses for real-time applications
  • long_context - Tasks requiring large context windows

use_cases string optional

Detailed description of ideal use cases for this model.

strengths string optional

Description of what this model excels at compared to other models.

limitations string optional

Known limitations or areas where this model may not perform optimally.

Performance Ratings

performance_ratings object optional

Comparative performance ratings for this model (1-5 scale, where 5 is best).

Show properties

reasoning integer optional

Reasoning capability rating (1-5). Indicates the model's ability to perform complex logical reasoning, multi-step problem solving, and analytical tasks.

speed integer optional

Response speed rating (1-5). Indicates how quickly the model generates responses. Higher values mean faster response times.

price_to_performance integer optional

Price-to-performance ratio rating (1-5). Indicates cost efficiency relative to capabilities. Higher values mean better value for money.

input_capabilities integer optional

Input modality support rating (1-5). Indicates variety and quality of supported input types (text, images, audio, etc.).

output_capabilities integer optional

Output modality support rating (1-5). Indicates variety and quality of supported output types and formats.

Performance & Benchmarks

benchmark_scores object optional

Objective performance benchmark scores for this model.

Show properties

mmlu number optional

MMLU (Massive Multitask Language Understanding) score (0-100).

humaneval number optional

HumanEval coding benchmark score (0-100).

gsm8k number optional

GSM8K math reasoning benchmark score (0-100).

hellaswag number optional

HellaSwag commonsense reasoning score (0-100).

Documentation

documentation_url string optional

URL to the detailed documentation page for this model.

provider_docs_url string optional

URL to the provider's official documentation for this model.

Display

badge string optional

Display badge for the model. One of new, beta, recommended, popular, fastest.

Usage

This object is returned by:

Returns

A ApiResponse object containing the API response data.

{
      "id": "mdl_09adacc3e5d3",
      "key": "gpt-4o",
      "name": "GPT-4o",
      "description": "Most advanced multimodal model with vision, audio, and tool calling capabilities",
      "provider": "openai",
      "is_visible_in_ui": true,
      "is_deprecated": false,
      "availability_status": "general_availability",
      "capabilities": [
        "text_generation",
        "conversation",
        "streaming",
        "tool_calling",
        "vision",
        "audio",
        "code_interpreter",
        "web_search",
        "file_search",
        "structured_output"
      ],
      "context_window": 128000,
      "pricing": {
        "input_synapses_cost": 2.50,
        "output_synapses_cost": 10.00
      },
      "release_date": "2024-05-13T00:00:00Z",
      "model_version": "2024-05-13",
      "training_data_cutoff": "2023-10-01T00:00:00Z",
      "added_to_aitronos_date": "2024-05-13T00:00:00Z",
      "recommended_for": [
        "general_purpose",
        "multimodal",
        "tool_calling",
        "vision",
        "audio"
      ],
      "use_cases": "Ideal for applications requiring multimodal understanding, complex reasoning, and tool integration. Excels at code generation, data analysis, image understanding, and audio processing.",
      "strengths": "Superior multimodal capabilities, advanced reasoning, excellent code generation, strong tool calling support, and high-quality vision understanding.",
      "limitations": "Higher cost compared to smaller models, may be overkill for simple text-only tasks.",
      "performance_ratings": {
        "reasoning": 4,
        "speed": 3,
        "price_to_performance": 3,
        "input_capabilities": 4,
        "output_capabilities": 4
      },
      "benchmark_scores": {
        "mmlu": 88.7,
        "humaneval": 92.0,
        "gsm8k": 94.8,
        "hellaswag": 95.3
      },
      "documentation_url": "https://docs.freddy.aitronos.com/models/gpt-4o",
      "provider_docs_url": "https://platform.openai.com/docs/models/gpt-4o",
      "badge": "recommended"
    }
Previous page
Next page