Presets API

A preset stores a default model, system prompt, sampling parameters and provider routing rules. Use it in any inference request as model: "@preset/<slug>".

Request

POST /api/v1/presets creates a preset in the authenticated workspace:

Other routes

PATCH update behavior: name is updated if provided and non-empty; description is tri-state (omit to keep, null to clear, string to set); config is shallow-merged so existing keys not in the patch are preserved.

PresetConfig fields

All fields are optional. A request can override these defaults.

Reasoning on anyrouter/* routes

The built-in anyrouter/auto, anyrouter/free, anyrouter/cowork, anyrouter/hermes and anyrouter/latest ids are routing presets, not fixed models, and have no synthetic reasoning level list. The rules below cover explicit request controls on them; a custom preset's stored reasoning object is not guaranteed to be merged by every endpoint.

• Chat requests use flat reasoning_effort. Responses uses nested reasoning.effort, with reasoning.enabled: false to disable. For flat Chat and converted Messages, none is a disable sentinel, not an effort tier.
• Messages requests use thinking.type (enabled / disabled); native Anthropic routes get the body unchanged. Converted to a non-native upstream, enabled maps to flat high and disabled to none; thinking.budget_tokens is not carried over.
• On flat reasoning_effort paths (including converted Messages), before each upstream attempt AnyRouter clamps the effort to the selected model's declared levels (a route-specific override wins), preferring the nearest lower level and falling back to the model default for an unknown tier. An omitted effort stays omitted. A branch's levels never leak into the next fallback branch. none is never clamped.
• If a resolved branch declares no effort enum, the caller's flat value is forwarded unchanged. For a selected route, its per-upstream supported_parameters is the authority.
• Preset-level require_params is a hard candidate constraint on paths that support preset candidate selection. Preset, key and request provider lists are unioned at the preference layer, but provider/key requirements are not a hard filter on every dispatch. Built-in virtual resolution uses tool-related entries (and request tools) to require tool-capable members; other parameters are not capability guarantees.
• provider.require_parameters: true is a separate strict model-level check over the request body and has no effect when the model declares no supported_parameters. Body-inferred parameters such as tools, tool_choice and response_format use the softer reorder/strip path. With no declared support list, support is treated as unknown and permissive.
• Routing floors merge as follows: min_context takes the higher value, key ignore is unioned, key and request only lists intersect. Top-level require_params is the legacy alias for provider.require_params; min_context is also accepted at the top level.
• A fallback chain may try a different model after an upstream failure; clamping is recalculated for that attempt.

Interleaved thinking is a provider beta request (anthropic-beta: interleaved-thinking-2025-05-14), not a preset or model guarantee. These rules do not promise toggle or token-budget support on every provider.

Response

GET /api/v1/presets returns a list envelope. A preset record:

{
  "id": "preset_abc123",
  "object": "preset",
  "slug": "code-reviewer",
  "name": "Code Reviewer",
  "description": "Strict senior-engineer review focused on bugs, security, and clarity.",
  "config": {
    "model": "anthropic/claude-sonnet-4",
    "system": "You are a senior code reviewer...",
    "temperature": 0.2,
    "max_tokens": 4096
  },
  "created_at": "2026-03-01T12:00:00.000Z",
  "updated_at": "2026-03-01T12:00:00.000Z"
}

DELETE /api/v1/presets/{slug} returns:

{
  "deleted": true,
  "slug": "code-reviewer"
}

Examples

curl -X POST https://anyrouter.dev/api/v1/presets \
  -H "Authorization: Bearer ak_your-management-key" \
  -H "Content-Type: application/json" \
  -d '{
    "slug": "code-reviewer",
    "name": "Code Reviewer",
    "description": "Strict senior-engineer review focused on bugs, security, and clarity.",
    "config": {
      "model": "anthropic/claude-sonnet-4",
      "system": "You are a senior code reviewer. Focus on bugs, security issues, and clarity.",
      "temperature": 0.2,
      "max_tokens": 4096
    }
  }'

import httpx

httpx.post(
    "https://anyrouter.dev/api/v1/presets",
    headers={"Authorization": "Bearer ak_your-management-key"},
    json={
        "slug": "code-reviewer",
        "name": "Code Reviewer",
        "config": {
            "model": "anthropic/claude-sonnet-4",
            "system": "You are a senior code reviewer.",
            "temperature": 0.2,
            "max_tokens": 4096,
        },
    },
)

await fetch("https://anyrouter.dev/api/v1/presets", {
  method: "POST",
  headers: {
    Authorization: "Bearer ak_your-management-key",
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    slug: "code-reviewer",
    name: "Code Reviewer",
    config: {
      model: "anthropic/claude-sonnet-4",
      system: "You are a senior code reviewer.",
      temperature: 0.2,
      max_tokens: 4096,
    },
  }),
})

Then set model to @preset/code-reviewer in any inference request.

Errors