Providers & Models
Model Catalog
The catalog of every model Multi-Router serves: identities, capabilities, limits, prices, lifecycle and updates.
For each model, the catalog records the identity, mode, capabilities and limits, and for each offering of the model it records the provider, variant, regions, model ids and prices. Multi-Router uses the catalog to check every request before calling the provider, to select the model id sent upstream, and to compute the cost of each request. The catalog is versioned and published by Multi-Router, and every installation serves one catalog version at a time.
Model entries
A model entry describes one model independently of where it is served. The Models page shows these fields in the Baseline panel of each model:
| Field | Description |
|---|---|
| Publisher | The company that built the model, such as Anthropic. |
| Slug | The model's unique name, such as anthropic/claude-sonnet-4.6. See Overview. |
| Product | The product line the model belongs to, such as Claude Sonnet or Qwen. |
| Version | The model version and its format, such as 4.6 · semver, or a date or snapshot id. |
| Released | The release date, when the publisher states one. |
| Mode | The type of model, such as Chat or Embedding. |
| Context | The context window and the maximum output tokens, such as 1,000,000 · 128,000 out. |
| Status | The lifecycle status: active, deprecated or retired. |
| Replaced by | The slug of the successor model, when the publisher names one. |
The Baseline panel also lists the model's capabilities, input modalities and aliases, with a link to the publisher's documentation.
Modes
The mode is the type of work a model performs. The Chat Completions and Messages endpoints serve chat models, and a request for a model of another mode returns wrong_endpoint_for_mode.
| Mode | Dashboard label | Models |
|---|---|---|
chat | Chat | Conversational and instruction models. |
responses | Responses | Models offered through the OpenAI Responses API. |
embedding | Embedding | Text embedding models. |
image_generation | Image gen | Image generation models. |
image_edit | Image edit | Image editing models. |
audio_transcription | Audio transcribe | Speech-to-text models. |
audio_speech | Audio speech | Text-to-speech models. |
realtime | Realtime | Low-latency audio and text conversation models. |
rerank | Rerank | Search reranking models. |
moderation | Moderation | Content moderation models. |
ocr | OCR | Document text extraction models. |
video_generation | Video gen | Video generation models. |
completion | Completion | Legacy text completion models. |
Capabilities and modalities
Capabilities are the request features a model supports, and modalities are the content types it accepts and produces, such as text, image, audio, video and PDF. Multi-Router checks every chat request against the model's capabilities and its image input modality, and rejects a request that uses a feature the model lacks before calling the provider:
| Check | Request feature | Error |
|---|---|---|
| Streaming | stream: true | streaming_not_supported |
| Tools | A non-empty tools list | tools_not_supported |
| Parallel tool calls | parallel_tool_calls: true | parallel_tool_calls_not_supported |
| Structured output | A json_object or json_schema response format | structured_output_not_supported |
| Reasoning | reasoning_effort or thinking | reasoning_not_supported |
| Prompt caching | prompt_cache_key or any cache_control | prompt_caching_not_supported |
| Image input modality | Image content in a message | modality_not_supported |
Offerings
An offering is one way to call a model: a provider and variant, with the regions where it is available. Each region carries the model id Multi-Router sends to the provider and the price. An offering can set limits or capabilities that differ from the model's baseline, such as a smaller context window on one provider, and an expanded offering on the Models page shows them as Capability overrides and as Context or Max output with the baseline value. The Offerings panel lists each offering's Vendor, Variant, Regions and Pricing, marks whether the variant is a fixed snapshot or a rolling version the provider updates, and marks custom offerings. Expanding an offering shows every region with its model id and prices. See Regions & Variants.
Prices
Prices are set for each region of each offering, in US dollars per million tokens unless noted, and are the providers' published list prices.
| Price | Description | Used in cost |
|---|---|---|
| Input | Prompt tokens that are not read from or written to the cache. | Yes |
| Output | Completion tokens, including reasoning tokens. | Yes |
| Cache read | Prompt tokens read from the prompt cache. | Yes |
| Cache write (5 minutes) | Prompt tokens written to the cache with a 5-minute lifetime. | Yes |
| Cache write (1 hour) | Prompt tokens written to the cache with a 1-hour lifetime. | Yes |
| Long context | Rates that replace all of the above when the prompt, including cached tokens, reaches a threshold such as 200,000 tokens. | Yes |
| Service tiers | Rates for the Batch, Priority, Flex and Fast tiers some providers offer, each with its own long context rates. The Fast tier applies when the provider reports that it served the request in fast mode. | Fast tier |
| Tool fees | Fees for provider-run tools, such as web search per 1,000 calls. | Web search |
| Cache storage | The cost of keeping one million cached tokens for one hour. | No |
| Modality rates | Input or output rates for audio, image or video content that differ from the token rates. | No |
| Unit prices | Prices per image, audio minute, page, video second, character, search or request, for models priced by unit. | No |
Multi-Router computes the cost of each request from the token counts, the number of web searches and the service tier the provider reports, and the price of the region the request was served in, and records it in Traffic, where it also counts toward budgets. A price part the catalog omits is charged at zero, and an offering whose provider publishes no price shows No pricing published on the Models page. Negotiated rates are applied with Custom Models & Pricing.
Lifecycle
The lifecycle status records whether a model is still offered. Each model has its own status, and each offering of a model can also have a status of its own:
| Status | Model | Offering |
|---|---|---|
| Active | Served normally. | Served normally. |
| Deprecated | Served, with the response headers x-mr-model-deprecated: true and, when the catalog names a successor, x-mr-model-replaced-by. | Served normally. |
| Retired | Refused with model_retired, whose message names the successor when there is one. | Refused with router_variant_retired for routers set to the retired variant. |
A model or offering with a scheduled retirement time becomes retired at that time, without a new catalog version. Claude Proxies serve active models only, and refuse a deprecated or retired model with HTTP 404 and a message that names its status.
Browsing the catalog
The Models page in the dashboard lists every model in the catalog, including models from providers the organization holds no key for, together with the organization's model groups and the installation's custom models.
| Control | Use |
|---|---|
| Pinned views | All models, Active, Reasoning, Vision and Tools. |
| My views | Saved views. Save view stores the current filters, search, status and Show choice. |
| Show | All, Models or Groups. |
| Status | All, Active or Deprecated. All includes retired models. |
| Filters | Publishers, Vendors, Capabilities (Vision, Audio, Video, Tools, Reasoning, Caching, Structured, Parallel tools, Streaming) and Mode (Chat only, All modes or a single mode). |
| Search | Matches the slug, name, publisher, aliases and provider model ids. |
| Columns | Kind, Publisher, Vendors, Capabilities, Context and Price, kept for each browser. |
The Price column shows the lowest-to-highest input and output prices across every region, a model priced by unit shows its unit price, and a model with no published price shows a dash. Selecting a model opens its Baseline and Offerings panels. In an expanded offering that publishes more than one rate, the Rate switch selects among the rates it publishes, such as Standard, Long Context, Batch, Priority, Flex, Fast and Tools.
Listing models with the API
GET /v1/models lists the chat models the API token's organization can call: models that are active or deprecated and have an offering that is not retired from a provider for which the organization holds an enabled provider key, together with custom endpoint models when the organization holds a Custom endpoint key. The endpoint requires the models:read scope and returns the OpenAI list format, sorted by publisher and slug. In each entry, owned_by is the publisher and created is the publication time of the catalog version in effect, which is the same for every model. The full catalog, with every model's offerings, regions and prices, is read with GET /v1/catalog/models and GET /v1/catalog/models/{publisher}/{model}, with the same scope; see Catalog.
With an x-mr-router header that names a Claude Proxy, the endpoint returns the models that Claude Proxy serves in the Anthropic format, which Claude Code uses to list its models, sorted by id. Each entry carries the model id the proxy's route uses, the display name, the release date at midnight UTC, and the token limits. A header that names any other kind of router returns 404.
Catalog updates
Multi-Router publishes the catalog as numbered versions at models.multi-router.ai, and a published version never changes. A new version adds models and offerings, updates prices, and records deprecations and retirements. Multi-Router manages catalog updates on the hosted service. On a self-hosted installation, the server administrator controls them on the Catalog tab of the Admin page:
| Control | Use |
|---|---|
| Catalog Updates | Turns checking the publisher on or off. While it is on, the installation checks every 30 seconds; turning it off keeps the version in effect. |
| Publisher URL | The address the catalog is fetched from, which can point to a private mirror. |
| Rollout mode | Auto applies each new version as it is fetched. Manual holds it for review, with Review diff, Approve go-live and Discard. |
| Refresh | Fetches the latest version immediately. |
| Recent versions | Lists earlier versions, with Roll back. Rolling back sets Rollout mode to Manual. |
Every fetched version is validated before use, a version that fails validation is skipped, and automatic updates only move forward. Upgrading the server installs the catalog bundled with the new release when it is newer than the version in effect. While an installation serves a catalog bundled with its release, the Models page shows a banner stating the version and that model data may be out of date. See Self-Hosting.
Errors
Catalog checks on chat requests return HTTP 400 with the following errors (503 for catalog_unavailable), in addition to the capability errors above. In a fallback router, a member that fails a catalog check is skipped and the next member is tried.
| Error | Cause |
|---|---|
model_not_found | The model is not in the catalog. |
wrong_endpoint_for_mode | The model is not a chat model. |
provider_not_offered | The router's provider does not offer the model. |
model_retired | The model is retired. The message names the successor when there is one. |
router_variant_missing | The router names no variant for the model. The message lists the valid variants. |
router_variant_not_found | The router's variant is not offered for the model. The message lists the valid variants. |
router_variant_retired | The router's variant is retired. The message lists the variants still served. |
publisher_not_supported | The router's provider is Google Vertex and the model is not published by Anthropic. |
region_not_served | The model is not offered in the selected region. |
catalog_unavailable | The catalog could not be read. Retry the request. |