Models API

Read the model catalog over REST. Each model has a provider/model slug, the same value as the chat completions model field. Use GET /api/v1/models (or https://api.anyrouter.dev/api/v1/models) for JSON; other /models URLs on some hosts serve HTML.

Request

List models

Returns every listed model to every caller, signed in or not. Access is a label, never a filter: each model says how it is reached (access) and whether this caller can route it now (can_route). Only your own choices remove a model: the key's allow/deny lists and your privacy blocks. Models the catalog hides or disables are not listed. The old all=1 and include=unavailable params from earlier clients are ignored: the list always has every model.

Invalid filter values (e.g. zdr=banana or a non-numeric price_max) are ignored, not errors.

Get one model

GET /api/v1/models/{owner}/{model} takes the slug as two path segments, for example /api/v1/models/anthropic/claude-sonnet-4-6. Nested routes (/routing-preview, /metrics, /privacy, /uptime) accept the slash form too; the encoded form (anthropic%2Fclaude-sonnet-4-6) also works.

Platform presets (anyrouter/auto, anyrouter/free, other first-party anyrouter/* ids) expose their live failover chain at GET /api/v1/models/anyrouter/auto/routing-preview, without auth, in the order inference tries it.

Count

GET /api/v1/models/count returns { "count": N }, the number of rows the list returns for the same query. It takes the same filters (provider, category, ...). Saved presets are not counted.

Response

{
  "object": "list",
  "data": [
    {
      "id": "openai/gpt-5.4-mini",
      "object": "model",
      "provider": "OpenAI",
      "providers": ["OpenAI", "Groq"],
      "category": "text",
      "context_length": 128000,
      "pricing": {
        "prompt": "10",
        "completion": "30",
        "input_per_1m": 10,
        "output_per_1m": 30
      },
      "capabilities": ["chat", "function-calling", "vision"],
      "links": {
        "details": "/api/v1/models/openai/gpt-4-turbo",
        "metrics": "/api/v1/models/openai%2Fgpt-4-turbo/metrics"
      }
    }
  ]
}

provider is the organization that owns the model. providers lists the public upstream providers that can serve it.

pricing.input_cache_read_per_1m is the cache-read rate per 1M tokens; omitted when the provider publishes none.

Who pays: pricing.billing

pricing.billing says how a request is paid for:

• "credits": a paid AnyRouter route serves the model and your credits pay for it.
• "free": a public $0 route serves it on the free tier.
• "byok": every route needs your own provider key. AnyRouter charges $0 and your provider bills you.

The default list can include "byok" models, so read the field. For them, pricing.list_pricing (input_per_1m, output_per_1m) holds the provider's published rate when known. A per_upstream entry priced at 0 only because it is BYOK carries billing: "byok" (plus list_pricing when published). Read 0 there as "billed by your provider", not free.

Access and routing: access, can_route

Every model carries three fields about how you can use it:

A model you cannot call is still in the list. Filter on can_route to show only models that work today:

{
  "id": "openai/gpt-6-cyber",
  "object": "model",
  "access": ["byok"],
  "can_route": false,
  "can_route_reason": "needs_key"
}

The same model for a caller with an OpenAI key saved in BYOK settings:

{
  "id": "openai/gpt-6-cyber",
  "object": "model",
  "access": ["byok"],
  "can_route": true
}

Reasoning and output limits

• reasoning.levels and reasoning.default are the model's effort enum when it declares one; omitted otherwise, so do not invent levels.
• per_upstream[].reasoning is a route-specific effort override; if omitted, the route inherits the model-level enum. per_upstream[].features.reasoning is a capability flag, not an effort list.
• supported_parameters is authoritative. For a selected route use per_upstream[].supported_parameters; the top-level field is a model-level projection, not a promise for every fallback route. Advertising reasoning does not add a reasoning parameter.
• top_provider.max_completion_tokens is the strictest completion cap declared by any serving upstream, or null when none declares one (not a promise of unlimited). Per-route caps are at per_upstream[].context.max_completion_tokens.
• AnyRouter does not infer toggle, token-budget or interleaved-thinking support from a model name. Toggle and budget controls are endpoint-specific (reasoning.enabled / reasoning.max_tokens on Responses, thinking on Messages). Anthropic interleaved thinking is a provider beta request (anthropic-beta: interleaved-thinking-2025-05-14), not a capability flag; support depends on the provider.

Your presets in the response

When the request is authenticated, the response also includes a presets array of your saved presets alongside data:

{
  "object": "list",
  "data": [ /* … models … */ ],
  "presets": [
    {
      "id": "preset_abc123",
      "object": "preset",
      "slug": "code-reviewer",
      "name": "Code Reviewer",
      "description": "Strict senior-engineer review.",
      "config": { "model": "anthropic/claude-sonnet-4-6", "temperature": 0.2 },
      "created_at": "2026-03-01T12:00:00Z",
      "updated_at": "2026-03-01T12:00:00Z"
    }
  ]
}

Omitted for unauthenticated requests; pass ?presets=0 to skip it when authenticated. Use a preset in any inference call as @preset/<slug>.

List providers

GET /api/v1/providers lists the upstream backends AnyRouter dispatches to:

{
  "object": "list",
  "data": [
    {
      "id": "openai",
      "name": "OpenAI",
      "description": "GPT models",
      "website": "https://openai.com",
      "status": "operational",
      "model_count": 4,
      "regions": ["us", "eu"],
      "model_prefix": "openai/",
      "endpoints": ["/api/v1/chat/completions", "/api/v1/responses", "/api/v1/models"]
    }
  ]
}

Examples

# Every model, anonymous. Use the api. host if apex returns HTML.
curl https://api.anyrouter.dev/api/v1/models

# The same list, labelled for this LLM API key (can_route, plus your presets)
curl https://anyrouter.dev/api/v1/models \
  -H "Authorization: Bearer sk-ar-your-key"

# Filter + newest first
curl "https://anyrouter.dev/api/v1/models?provider=anthropic&min_context=200000&sort=added" \
  -H "Authorization: Bearer sk-ar-your-key"

# Top usage this week
curl "https://anyrouter.dev/api/v1/models?sort=usage"

# One model (slash path)
curl https://anyrouter.dev/api/v1/models/anthropic/claude-sonnet-4-6 \
  -H "Authorization: Bearer sk-ar-your-key"

# Models with a public ZDR upstream
curl "https://anyrouter.dev/api/v1/models?zdr=true" \
  -H "Authorization: Bearer sk-ar-your-key"

# Free-tier platform routes only
curl "https://anyrouter.dev/api/v1/models?free=true" \
  -H "Authorization: Bearer sk-ar-your-key"

# Live failover chain for a platform preset (slash path; %2F also works)
curl https://anyrouter.dev/api/v1/models/anyrouter/auto/routing-preview

# List providers (no auth required)
curl https://anyrouter.dev/api/v1/providers

import requests

resp = requests.get(
    "https://anyrouter.dev/api/v1/models",
    params={"provider": "anthropic", "min_context": 200000},
    headers={"Authorization": "Bearer sk-ar-your-key"},
)
for model in resp.json()["data"]:
    print(model["id"], model["context_length"])

const res = await fetch(
  "https://anyrouter.dev/api/v1/models?provider=anthropic&min_context=200000",
  { headers: { Authorization: "Bearer sk-ar-your-key" } },
)
const { data } = await res.json()
data.forEach((m: { id: string; context_length: number }) =>
  console.log(m.id, m.context_length),
)

Errors

See Errors for the full status table.