Info

List available models

get

List all available models with restrictions per tier.

Returns models grouped by: mode -> source -> [models with tags] Plus model_restrictions_per_tier showing which models are restricted for each tier.

Modes:

  • auto: All available models (union of direct + opengpu). Use mode: "auto" for intelligent routing.

  • direct: Low-latency models via direct providers

  • opengpu: Decentralized models via blockchain

Tags:

  • text-to-text: LLM chat models (gpt-oss, llama, deepseek)

  • text-to-speech: TTS models (sesame)

  • speech-to-text: ASR models (whisper)

  • text-to-image: Image generation models (Stable Diffusion)

  • text-to-video: Video generation models (Wan, Sora)

Example response:

{
  "auto": {
    "ollama": [
      {"name": "gpt-oss:120b", "tag": "text-to-text"},
      {"name": "gpt-oss:20b", "tag": "text-to-text"}
    ],
    "audio": [
      {"name": "sesame/csm-1b", "tag": "text-to-speech"}
    ]
  },
  "direct": {...},
  "opengpu": {...},
  "model_restrictions_per_tier": {
    "guest": ["direct.video.openai/sora-2-T2V", ...],
    "basic": [],
    "pro": [],
    "max": []
  }
}
Responses
chevron-right
200

Successful Response

application/json
get
/v2/models
200

Successful Response

List available tiers

get

List all available tiers with their model restrictions and rate limits.

Returns:

  • All tier definitions: guest, basic, pro, max (sorted by tier level)

  • Model restrictions for each tier (blacklist - models in list are blocked)

  • Rate limits per mode (opengpu, direct, global RPM)

  • Current user's tier (guest if not authenticated)

Useful for:

  • Showing upgrade options in dashboard

  • Comparing tier features and restrictions

  • Understanding which models are blocked per tier

Header parameters
x-api-keyany ofOptional
stringOptional
or
nullOptional
Responses
chevron-right
200

Successful Response

application/json
get
/v2/tiers

List model pricing

get

List pricing for all available models.

Returns per-request cost in credits for each mode/model combination.

Note: Currently only per-request pricing is shown. Token-based pricing may be added in the future.

Example response:

{
  "pricing": [
    {"mode": "direct", "model": "ollama.gpt-oss:120b", "per_request": 0.001},
    {"mode": "direct", "model": "video.Wan-AI/Wan2.5-T2V", "per_request": 0.01},
    {"mode": "opengpu", "model": "ollama.gpt-oss:20b", "per_request": 0.00025}
  ],
  "total_count": 3
}
Responses
chevron-right
200

Successful Response

application/json
get
/v2/pricing
200

Successful Response

Last updated