Model Management

Discover and query available AI models and their capabilities across all providers.

List Models

Returns the complete list of currently enabled AI models with their capabilities and limits.

Endpoint: https://apis.threatwinds.com/api/ai/v1/models

Method: GET

Parameters

Headers

Header	Type	Required	Description
Authorization	string	Optional*	Bearer token for session authentication
api-key	string	Optional*	API key for key-based authentication
api-secret	string	Optional*	API secret for key-based authentication

Note: You must use either Authorization header OR API key/secret combination.

Request

curl -X 'GET' \
  'https://apis.threatwinds.com/api/ai/v1/models' \
  -H 'accept: application/json' \
  -H 'Authorization: Bearer <YOUR_BEARER_TOKEN>'

Response

Success Response (200 OK)

{
  "object": "list",
  "data": [
    {
      "id": "silas-1.6",
      "object": "model",
      "name": "Silas 1.6",
      "provider": "threatwinds",
      "owned_by": "ThreatWinds",
      "created": 2026,
      "capabilities": [
        "chat",
        "image",
        "video",
        "tools-use",
        "text-generation",
        "code-generation",
        "reasoning"
      ],
      "limits": {
        "max_input_tokens": 868928,
        "max_total_tokens": 1000000
      },
      "params": {
        "temperature": 0.6,
        "temperature_min": 0.0,
        "temperature_max": 1.0,
        "top_p": 0.95,
        "top_k": 20,
        "min_p": 0.0,
        "presence_penalty": 0.0,
        "frequency_penalty": 0.0,
        "repetition_penalty": 1.0,
        "max_completion_tokens": 131072
      }
    }
  ]
}

Response Schema

Field	Type	Description
object	string	Response type, always “list”
data	array	Array of model objects
data[].id	string	Unique model identifier
data[].name	string	Human-readable model name
data[].provider	string	Provider ID: `threatwinds` for all self-hosted models (chat, embeddings, audio); `openai`, `claude`, or `gemini` for third-party models
data[].owned_by	string	Organization or developer
data[].created	integer	Year of release
data[].capabilities	array	Supported features
data[].limits	object	Token constraints
limits.max_input_tokens	integer	Maximum input context size
limits.max_total_tokens	integer	Combined input+output limit
data[].embeddings	object	Embedding model configuration (embedding models only). Contains `default_dimensions`, `min_dimensions`, `max_dimensions`
params	object	Model parameters and validation ranges (optional)

Capability Values

Value	Description
chat	Conversational text interaction
text-generation	General-purpose text generation
code-generation	Programming code creation and completion
tools-use	Function calling capability
reasoning	Step-by-step reasoning support
image	Vision/image understanding
video	Video understanding
transcription	Speech-to-text
speech	Text-to-speech synthesis
embeddings	Vector embedding generation
vision-embeddings	Multimodal embeddings

Note: Additional capability values may include audio, audio-generation, image-generation, video-generation.

Error Codes

Status Code	Description	Cause
200	OK	Success
400	Bad Request	Invalid parameters
401	Unauthorized	Missing/invalid credentials
403	Forbidden	Insufficient permissions

Get Model Details

Retrieve detailed information about a specific model by ID.

Endpoint: https://apis.threatwinds.com/api/ai/v1/models/{id}

Method: GET

Path Parameters

Parameter	Type	Required	Description	Example
id	string	Yes	Model identifier	silas-1.6

Request

curl -X 'GET' \
  'https://apis.threatwinds.com/api/ai/v1/models/silas-1.7-pro' \
  -H 'Authorization: Bearer <YOUR_BEARER_TOKEN>'

Response

{
  "id": "silas-1.7-pro",
  "object": "model",
  "name": "Silas 1.7 Pro",
  "provider": "threatwinds",
  "owned_by": "ThreatWinds",
  "created": 2026,
  "capabilities": ["chat", "image", "video", "tools-use", "text-generation", "code-generation", "reasoning"],
  "limits": {
    "max_input_tokens": 868928,
    "max_total_tokens": 1000000
  },
  "params": {
    "temperature": 1.0,
    "top_p": 0.95,
    "top_k": 20,
    "min_p": 0.0,
    "presence_penalty": 0.0,
    "frequency_penalty": 0.0,
    "repetition_penalty": 1.0,
    "max_completion_tokens": 131072
  }
}

Error Codes

Status Code	Description	Cause
200	OK	Model details returned successfully
404	Not Found	The specified model ID does not exist in any provider configuration

Filtering Models Programmatically

Use jq or similar tools to filter the /models response for your needs.

Filter by Provider

Get all models from a specific provider:

# Self-hosted ThreatWinds models
curl -s 'https://apis.threatwinds.com/api/ai/v1/models' \
  -H 'Authorization: Bearer <YOUR_BEARER_TOKEN>' | \
  jq '.data[] | select(.provider == "threatwinds")'

# OpenAI models
curl -s 'https://apis.threatwinds.com/api/ai/v1/models' \
  -H 'Authorization: Bearer <YOUR_BEARER_TOKEN>' | \
  jq '.data[] | select(.provider == "openai")'

Filter by Capability

Find models supporting specific features:

# Models with vision/image capability
curl -s 'https://apis.threatwinds.com/api/ai/v1/models' \
  -H 'Authorization: Bearer <YOUR_BEARER_TOKEN>' | \
  jq '.data[] | select(.capabilities | contains(["image"]))'

# Models supporting tool calls  
curl -s 'https://apis.threatwinds.com/api/ai/v1/models' \
  -H 'Authorization: Bearer <YOUR_BEARER_TOKEN>' | \
  jq '.data[] | select(.capabilities | contains(["tools-use"]))'

# Embedding models only
curl -s 'https://apis.threatwinds.com/api/ai/v1/models' \
  -H 'Authorization: Bearer <YOUR_BEARER_TOKEN>' | \
  jq '.data[] | select(.capabilities | contains(["embeddings"]))'

# Audio transcription models
curl -s 'https://apis.threatwinds.com/api/ai/v1/models' \
  -H 'Authorization: Bearer <YOUR_BEARER_TOKEN>' | \
  jq '.data[] | select(.capabilities | contains(["transcription"]))'

Filter by Context Window

Models with large context windows:

# Models with > 100K input tokens
curl -s 'https://apis.threatwinds.com/api/ai/v1/models' \
  -H 'Authorization: Bearer <YOUR_BEARER_TOKEN>' | \
  jq '.data[] | select(.limits.max_input_tokens > 100000)'

# Sort by context size (largest first)
curl -s 'https://apis.threatwinds.com/api/ai/v1/models' \
  -H 'Authorization: Bearer <YOUR_BEARER_TOKEN>' | \
  jq '.data | sort_by(-.limits.max_input_tokens) | .[0:5]'

Extract Useful Fields

Create clean lists for UI or CLI usage:

# Just model IDs
curl -s 'https://apis.threatwinds.com/api/ai/v1/models' \
  -H 'Authorization: Bearer <YOUR_BEARER_TOKEN>' | \
  jq -r '.data[].id'

# Display name and provider
curl -s 'https://apis.threatwinds.com/api/ai/v1/models' \
  -H 'Authorization: Bearer <YOUR_BEARER_TOKEN>' | \
  jq -r '.data[] | "\(.name) (\(.provider))"'

# Chat-capable models with context info
curl -s 'https://apis.threatwinds.com/api/ai/v1/models' \
  -H 'Authorization: Bearer <YOUR_BEARER_TOKEN>' | \
  jq -r '.data[] | select(.capabilities | contains(["chat"])) | "\(.id): \(.limits.max_input_tokens) tokens"'

Model Discovery Best Practices

Query at Runtime: Call /models during application initialization to get current catalog
Cache Appropriately: Cache model listings but refresh periodically as providers add/deprecate models
Check Capabilities First: Always verify .capabilities array before using a model for specific tasks
Respect Token Limits: Read .limits to ensure your use case fits within model constraints
Handle Unknown Errors: Some providers may have temporary outages; implement retry logic
Provider-Specific Behavior: Check params for valid parameter ranges (e.g., temperature bounds)

Provider Overview

Provider	Type	Models
threatwinds	Self-hosted	Silas chat (`silas-1.6`, `silas-1.6-pro`, `silas-1.7-pro`), text/vision embeddings (`qwen3-embedding-8b`, `qwen3-vl-embedding-8b`), and audio (`whisper-large-v3`, `kokoro-82m`). Use the `owned_by` field to identify the original model maintainer.
openai	External API	GPT-6 ladder: `gpt-6-astra`, `gpt-6.1-sol`, `gpt-6-luna`, plus the legacy aliases `gpt-5-nano` and `gpt-5.4-nano` (both route to `gpt-6-luna`).
claude	External API	Claude 5 series: `claude-opus-5`, `claude-sonnet-5`, `claude-fable-5`, `claude-fable-5-1`.
gemini	External API	`gemini-3.1-pro`, `gemini-3.8-flash`, `gemini-3.5-flash-lite`, and the embedding model `gemini-embedding-2`.

Note: The active model set varies based on deployment configuration. Always check /models for real-time availability.