Residency & models
The response disclosure headers, managed aliases, and the live model catalog with availability and fallback notes.

Response headers
Every call routed to a concrete model returns x-sluis-model, the provider/model that served it, residency reroutes included. Calls to a managed sluis/* alias also return x-sluis-route. Any call that carried an inline document also discloses how that document reached the model.
Residency itself is organisation policy, enforced at dispatch, never a request header.
Gateway refusals that ask you to wait carry Retry-After: a 429 rate-limit refusal gives the whole seconds until the next per-minute window (1 to 60), a 503 for unavailable billing admission gives 1.
| Header | Description |
|---|---|
| x-sluis-route | Response, on managed alias calls: the route key that resolved the request, for example auto for sluis/auto. Not sent for tenant aliases. |
| x-sluis-model | Response, on every call routed to a concrete model, not only aliases: the provider/model that actually served it after policy and routing. |
| x-sluis-document-route | Response, on any call that carried an inline PDF or Word document: how it reached the model. images (one image per page), text (extracted text only, the pages themselves were not sent), native (forwarded to a provider that reads the document itself) or mixed. Absent when the request carried no document. |
| x-sluis-unscanned | Response, when part of the request could not be inspected by data protection, for example audio. The value names the part type, such as audio. |
Cost of a call
Send x-sluis-include-cost: true (true or 1, any case) to see what one call cost your organisation. Any other value switches it off and is never an error. Without the header every response is unchanged.
The price is in EUR, excluding VAT, and is the amount the call books on your invoice or draws from your prepaid balance: the provider list price plus the Sluis fee after your organisation's fee discount, times an agency margin if you are a client organisation. A call on your own provider key or a custom provider costs the flat EUR 0.50 per million input plus output tokens, with the same discount and margin. Discount and margin are not itemised.
- Non-streaming: the response carries
x-sluis-cost-eur, the price with exactly six decimals, for example0.000420.x-sluis-cost-estimated: trueis added only when usage was estimated or the price imputed. - Streaming: the final usage object carries
sluis_cost_euras a JSON number, for example0.00042: in the last chunk of OpenAI-shaped streams (ask forstream_options.include_usage) and in themessage_deltaevent of Anthropic streams.sluis_cost_estimated: trueappears only when estimated. A stream that ends without a usage frame gets nothing injected; the estimate is still billed. - Cache: an exact-cache hit always carries
x-sluis-cache: hitand costs nothing, so with the opt-inx-sluis-cost-euris0.000000. - With the OpenAI Python SDK read headers through the raw response:
client.chat.completions.with_raw_response.create(..., extra_headers={"x-sluis-include-cost": "true"}), thenraw.headers["x-sluis-cost-eur"]andraw.parse()for the usual object. - With curl,
-D -prints the response headers.
curl -sS -D - https://api.sluis.ai/v1/chat/completions \ -H "Authorization: Bearer $SLUIS_KEY" \ -H "x-sluis-include-cost: true" \ -H "Content-Type: application/json" \ -d '{ "model": "mistral/mistral-large-latest", "messages": [{ "role": "user", "content": "Say hi" }] }' # response headers: # x-sluis-model: mistral/mistral-large-latest # x-sluis-cost-eur: 0.000420
raw = client.chat.completions.with_raw_response.create(
model="mistral/mistral-large-latest",
messages=[{"role": "user", "content": "Say hi"}],
extra_headers={"x-sluis-include-cost": "true"},
)
print(raw.headers["x-sluis-cost-eur"]) # 0.000420
completion = raw.parse() # the usual ChatCompletion object# "stream": true with stream_options.include_usage: the final chunk data: {"object":"chat.completion.chunk","choices":[{"index":0,"delta":{},"finish_reason":"stop"}],"usage":{"prompt_tokens":12,"completion_tokens":9,"total_tokens":21,"sluis_cost_eur":0.000042}} # Anthropic /v1/messages: the same field in the message_delta event's usage data: {"type":"message_delta","delta":{"stop_reason":"end_turn","stop_sequence":null},"usage":{"output_tokens":8,"sluis_cost_eur":0.000042}}
Managed aliases
Managed aliases are reserved sluis/... model names. sluis/auto chooses per request among the routes your policy allows. The use-case aliases choose from daily Artificial Analysis benchmark snapshots curated by Sluis.
Every alias resolves inside the organisation's residency policy first, so the sovereignty rule beats the benchmark rank.
How automatic routing chooses
- A classifier inside the Sluis deployment can pick the reasoning route without any text leaving it.
- Otherwise a hosted classifier reads three short excerpts of the prompt after data protection. It only ever runs on a model from an EU-owned provider, processing in the EU.

| Alias | Use case |
|---|---|
| sluis/ | Per-request routing. Sluis picks the best compliant target for the prompt. |
| sluis/ | Coding and code review. |
| sluis/ | General conversation. |
| sluis/ | Customer-support answers. |
| sluis/ | Low latency. |
| sluis/ | Best value for bulk work. |
| sluis/ | Tool use and agentic workflows. |
| sluis/ | Structured extraction. |
| sluis/ | Document extraction and parsing. |
| sluis/ | Long-context analysis. |
| sluis/ | Image input. |
| sluis/ | Multilingual translation. |
| sluis/ | Open-weights only. |
| sluis/ | Math, science and multi-step reasoning. |
Call an alias anywhere you would pass a provider/model id. The response headers show both the route you asked for and the concrete model that ran.
curl https://api.sluis.ai/v1/chat/completions \ -H "Authorization: Bearer $SLUIS_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "sluis/auto", "messages": [{ "role": "user", "content": "Plan a GDPR-safe rollout" }] }' # response headers: # x-sluis-route: auto # x-sluis-model: mistral/mistral-large-latest
x-sluis-modelis the concreteprovider/modeltarget after residency policy, shadowing and routing.x-sluis-routeis the route key of a managed alias. Tenant aliases do not set it.- If the provider refuses the credential for a managed route's model, the call moves on to the route's next compliant model. A concrete
provider/modelor a tenant alias is never swapped.
Tenant aliases shadow managed names
If your organisation creates sluis/code, that route wins for your tenant.
Workload allow-lists may include reserved aliases such as sluis/auto. The allow-list is checked before the alias resolves.
Supported models
Sluis exposes built-in providers through managed keys when a platform credential is configured. To override or extend the catalog, connect your own key (BYOK) or add a custom OpenAI-compatible provider in the Console.
Every callable id is provider-prefixed. Pass provider/model (for example mistral/mistral-large-latest) and Sluis routes it per your residency policy.
GET /v1/modelsreturnsobject: listanddata;GET /v1/models/{id}returns one model object (404when the model is not in this key's catalog). Every object has the same fields:id,object,owned_by,created: 1704067200(a fixed compatibility timestamp, not a release date),kind,context_windowin tokens,capabilities,residency,resolves_toandpricing.kindistext,multimodal,embedding,decision,ocroraudio.capabilitiesalways hastools,visionandreasoning.residencyhasjurisdiction,region,countryandowner_is_eu.resolves_tois a tenant alias's concrete target; the alias repeats its target's metadata and price. Unknown values arenull, never false or zero; a knownfalsecapability means unsupported.- Managed
sluis/*routes choose the model per request: theirkind,context_window,residencyandpricingarenull, and every capability isnull.sluis/ocr,sluis/transcribeandsluis/typed/routerhaveresidency: null. pricingis the final price this organisation pays, in EUR excluding VAT, rounded to six decimals.input_eur_per_mtok,output_eur_per_mtokandcached_input_eur_per_mtokare per million tokens.audio_input_eur_per_mtokandaudio_output_eur_per_mtokare per million audio tokens and are set only for a model whose audio tokens are priced apart from its text tokens, such asvertex/gemini-3.8-liveand audio-capable chat models where the provider lists an audio rate; otherwise they arenulland audio bills at the token rates.long_contextlists steps withfrom_tokensand input and output rates that apply to the whole call.eur_per_page,eur_per_minuteandeur_per_imagecover other units; a rate that does not apply isnull.estimatedistruewhen the price book has no entry and metering imputes a rate.- How the price is built: provider list price plus Sluis's 10% fee, with the negotiated discount applied to that fee only, then multiplied by any agency margin. A list rate of EUR 1.00 with a 20% discount and a 5% margin lists as 1.00 × 1.08 × 1.05 = 1.134. On the cloud, your own key on a built-in provider and custom providers cost the flat EUR 0.50 per million input and output tokens, times (1 - discount) × (1 + margin); a declared custom price does not lower this fee and your provider charges separately. Only the final price is returned, not its discount or margin components.
- On a licensed appliance, custom-provider prices are the declared token rates, or zero without a declaration; your own key on a built-in provider is listed and billed like an included model.
- A changed fee discount or margin shows within about 30 seconds. The price book refreshes about every six hours, so listed prices can still change; discovery is not a locked quote. Prices are computed in memory and add no provider call or database read. Anthropic-shaped responses selected by
anthropic-versionor/anthropic/v1/modelsare unchanged.
How the catalog is built
- The catalog is credential- and policy-aware.
- It uses live provider catalogs where available, and can fall back to declared models during an upstream catalog failure.
- Vertex candidates are probe-verified separately for EU availability.
- Nebius Token Factory is EU-owned, but its public
nebius/...models have no region guarantee. They count as GLOBAL and need an explicit GLOBAL opt-in. - Only
nebius-eu, the operator's dedicated endpoints in an EU region, counts as EU.

Live catalog. This public platform catalog follows the gateway's ten-minute cache and is also available at api.sluis.ai/public/models.json. Authenticated GET /v1/models applies your policy and workload allow-list and returns this organisation's final EUR prices, excluding VAT. The price book refreshes about every six hours; listed prices can change.
| anthropic | Region | Context | tok/s |
|---|---|---|---|
| anthropic/ | US · United States | 1M | n/a |
| anthropic/ | US · United States | 1M | n/a |
| anthropic/ | US · United States | 1M | n/a |
| anthropic/ | US · United States | 200k | n/a |
| anthropic/ | US · United States | 1M | 235† |
| anthropic/ | US · United States | 200k | n/a |
| anthropic/ | US · United States | 1M | n/a |
| anthropic/ | US · United States | 1M | 50† |
| anthropic/ | US · United States | 1M | 60† |
| anthropic/ | US · United States | 1M | 53† |
| anthropic/ | US · United States | 1M | n/a |
| anthropic/ | US · United States | 200k | n/a |
| anthropic/ | US · United States | 1M | n/a |
| anthropic/ | US · United States | 1M | 85† |
| anthropic/ | US · United States | 1M | 132‡ |
| bedrock | Region | Context | tok/s |
|---|---|---|---|
| bedrock/ | EU · EU (cross-region) | 1M | n/a |
| bedrock/ | EU · EU (cross-region) | 300k | 177† |
| bedrock/ | EU · EU (cross-region) | 128k | 275† |
| bedrock/ | EU · EU (cross-region) | 300k | 135† |
| bedrock/ | EU · EU (cross-region) | n/a | n/a |
| bedrock/ | EU · EU (cross-region) | n/a | n/a |
| bedrock/ | EU · EU (cross-region) | 200k | n/a |
| bedrock/ | EU · EU (cross-region) | 1M | 235† |
| bedrock/ | EU · EU (cross-region) | 200k | n/a |
| bedrock/ | EU · EU (cross-region) | 1M | n/a |
| bedrock/ | EU · EU (cross-region) | 1M | 50† |
| bedrock/ | EU · EU (cross-region) | 1M | 60† |
| bedrock/ | EU · EU (cross-region) | 1M | 53† |
| bedrock/ | EU · EU (cross-region) | 1M | n/a |
| bedrock/ | EU · EU (cross-region) | 200k | n/a |
| bedrock/ | EU · EU (cross-region) | 1M | n/a |
| bedrock/ | EU · EU (cross-region) | 1M | 37‡ |
| bedrock/ | EU · EU (cross-region) | 1M | n/a |
| bedrock/ | EU · EU (cross-region) | 128k | n/a |
| bedrock/ | EU · EU (cross-region) | 128k | n/a |
| bedrock/ | EU · EU (cross-region) | 128k | n/a |
| bedrock/ | EU · EU (cross-region) | 128k | n/a |
| mistral | Region | Context | tok/s |
|---|---|---|---|
| mistral/ | EU · Paris, France | 128k | n/a |
| mistral/ | EU · Paris, France | 8k | n/a |
| mistral/ | EU · Paris, France | 8k | n/a |
| mistral/ | EU · Paris, France | 128k | n/a |
| mistral/ | EU · Paris, France | 1M | 173† |
| mistral/ | EU · Paris, France | 1M | 72† |
| mistral/ | EU · Paris, France | n/a | n/a |
| mistral/ | EU · Paris, France | 262k | n/a |
| mistral/ | EU · Paris, France | 262k | n/a |
| mistral/ | EU · Paris, France | 262k | n/a |
| mistral/ | EU · Paris, France | 262k | n/a |
| mistral/ | EU · Paris, France | 262k | n/a |
| mistral/ | EU · Paris, France | 262k | n/a |
| mistral/ | EU · Paris, France | 131k | n/a |
| mistral/ | EU · Paris, France | 131k | n/a |
| mistral/ | EU · Paris, France | 262k | n/a |
| mistral/ | EU · Paris, France | 262k | n/a |
| mistral/ | EU · Paris, France | 128k | n/a |
| mistral/ | EU · Paris, France | 128k | n/a |
| mistral/ | EU · Paris, France | 8k | n/a |
| mistral/ | EU · Paris, France | 8k | n/a |
| mistral/ | EU · Paris, France | 262k | n/a |
| mistral/ | EU · Paris, France | 1M | 95† |
| mistral/ | EU · Paris, France | 1M | n/a |
| mistral/ | EU · Paris, France | 262k | 42‡ |
| mistral/ | EU · Paris, France | 262k | n/a |
| mistral/ | EU · Paris, France | 262k | n/a |
| mistral/ | EU · Paris, France | 262k | 46† |
| mistral/ | EU · Paris, France | 262k | 173† |
| mistral/ | EU · Paris, France | 262k | 173† |
| mistral/ | EU · Paris, France | 262k | n/a |
| mistral/ | EU · Paris, France | 131k | n/a |
| mistral/ | EU · Paris, France | n/a | n/a |
| mistral/ | EU · Paris, France | n/a | n/a |
| mistral/ | EU · Paris, France | n/a | n/a |
| mistral/ | EU · Paris, France | n/a | n/a |
| mistral/ | EU · Paris, France | n/a | n/a |
| mistral/ | EU · Paris, France | n/a | n/a |
| mistral/ | EU · Paris, France | 262k | n/a |
| mistral/ | EU · Paris, France | 262k | n/a |
| mistral/ | EU · Paris, France | 262k | n/a |
| mistral/ | EU · Paris, France | 262k | n/a |
| mistral/ | EU · Paris, France | 262k | n/a |
| mistral/ | EU · Paris, France | n/a | n/a |
| mistral/ | EU · Paris, France | n/a | n/a |
| mistral/ | EU · Paris, France | n/a | n/a |
| mistral/ | EU · Paris, France | n/a | n/a |
| mistral/ | EU · Paris, France | n/a | n/a |
| mistral/ | EU · Paris, France | n/a | n/a |
| mistral/ | EU · Paris, France | n/a | n/a |
| mistral/ | EU · Paris, France | n/a | n/a |
| mistral/ | EU · Paris, France | n/a | n/a |
| mistral/ | EU · Paris, France | 32k | n/a |
| mistral/ | EU · Paris, France | 32k | n/a |
| mistral/ | EU · Paris, France | n/a | n/a |
| mistral/ | EU · Paris, France | n/a | n/a |
| mistral/ | EU · Paris, France | 1M | n/a |
| mistral/ | EU · Paris, France | 1M | n/a |
| mistral/ | EU · Paris, France | 1M | n/a |
| mistral/ | EU · Paris, France | 1M | n/a |
| mistral/ | EU · Paris, France | 1M | n/a |
| moonshot | Region | Context | tok/s |
|---|---|---|---|
| moonshot/ | CN · China | 262k | n/a |
| moonshot/ | CN · China | 262k | 61† |
| moonshot/ | CN · China | n/a | n/a |
| moonshot/ | CN · China | 1M | 56† |
| moonshot/ | CN · China | 1M | 29† |
| moonshot/ | CN · China | 1M | n/a |
| nebius | Region | Context | tok/s |
|---|---|---|---|
| nebius/ | GLOBAL · Worldwide (no region guarantee) | 1M | 124† |
| nebius/ | GLOBAL · Worldwide (no region guarantee) | 1M | 162† |
| nebius/ | GLOBAL · Worldwide (no region guarantee) | 131k | n/a |
| nebius/ | GLOBAL · Worldwide (no region guarantee) | 262k | 46† |
| nebius/ | GLOBAL · Worldwide (no region guarantee) | 262k | n/a |
| nebius/ | GLOBAL · Worldwide (no region guarantee) | 40k | n/a |
| nebius/ | GLOBAL · Worldwide (no region guarantee) | 1M | 45† |
| nebius/ | GLOBAL · Worldwide (no region guarantee) | 1M | 95† |
| nebius/ | GLOBAL · Worldwide (no region guarantee) | 1M | 173† |
| nebius/ | GLOBAL · Worldwide (no region guarantee) | 110k | n/a |
| nebius/ | GLOBAL · Worldwide (no region guarantee) | 1M | 221† |
| nebius/ | GLOBAL · Worldwide (no region guarantee) | 1M | 56† |
| nebius/ | GLOBAL · Worldwide (no region guarantee) | 1M | 29† |
| nebius/ | GLOBAL · Worldwide (no region guarantee) | 262k | n/a |
| nebius/ | GLOBAL · Worldwide (no region guarantee) | 1M | n/a |
| nebius/ | GLOBAL · Worldwide (no region guarantee) | 1M | 275† |
| nebius/ | GLOBAL · Worldwide (no region guarantee) | 1M | 290† |
| nebius/ | GLOBAL · Worldwide (no region guarantee) | 262k | n/a |
| nebius/ | GLOBAL · Worldwide (no region guarantee) | 131k | 228† |
| nebius/ | GLOBAL · Worldwide (no region guarantee) | 1M | n/a |
| nebius/ | GLOBAL · Worldwide (no region guarantee) | 1M | 111† |
| nebius/ | GLOBAL · Worldwide (no region guarantee) | 1M | 72† |
| nebius/ | GLOBAL · Worldwide (no region guarantee) | 1M | 103† |
| nebius/ | GLOBAL · Worldwide (no region guarantee) | 1M | 61† |
| nebul | Region | Context | tok/s |
|---|---|---|---|
| nebul/ | EU · European Union | 8k | n/a |
| nebul/ | EU · European Union | 8k | n/a |
| nebul/ | EU · European Union | 32k | n/a |
| nebul/ | EU · European Union | 40k | n/a |
| nebul/ | EU · European Union | 262k | n/a |
| nebul/ | EU · European Union | 12k | n/a |
| nebul/ | EU · European Union | 262k | n/a |
| nebul/ | EU · European Union | 4k | n/a |
| nebul/ | EU · European Union | 512 | n/a |
| nebul/ | EU · European Union | 8k | n/a |
| nebul/ | EU · European Union | 512 | n/a |
| nebul/ | EU · European Union | 8k | n/a |
| nebul/ | EU · European Union | 4k | n/a |
| nebul/ | EU · European Union | 262k | n/a |
| nebul/ | EU · European Union | 262k | n/a |
| nebul/ | EU · European Union | 262k | n/a |
| nebul/ | EU · European Union | 1M | n/a |
| nebul/ | EU · European Union | 4k | n/a |
| nebul/ | EU · European Union | 4k | n/a |
| nebul/ | EU · European Union | 131k | 228† |
| nebul/ | EU · European Union | n/a | n/a |
| nebul/ | EU · European Union | 512 | n/a |
| nebul/ | EU · European Union | 1M | 145‡ |
| nebul/ | EU · European Union | 8k | n/a |
| openai | Region | Context | tok/s |
|---|---|---|---|
| openai/ | US · United States | 400k | n/a |
| openai/ | US · United States | n/a | n/a |
| openai/ | US · United States | 1M | 183† |
| openai/ | US · United States | 1M | 183† |
| openai/ | US · United States | 1M | 150† |
| openai/ | US · United States | 1M | 150† |
| openai/ | US · United States | 128k | 197† |
| openai/ | US · United States | 128k | 197† |
| openai/ | US · United States | 128k | 197† |
| openai/ | US · United States | 128k | 81‡ |
| openai/ | US · United States | 128k | 56† |
| openai/ | US · United States | 16k | n/a |
| openai/ | US · United States | 16k | n/a |
| openai/ | US · United States | 16k | n/a |
| openai/ | US · United States | n/a | n/a |
| openai/ | US · United States | n/a | n/a |
| openai/ | US · United States | n/a | n/a |
| openai/ | US · United States | 16k | n/a |
| openai/ | US · United States | 16k | n/a |
| openai/ | US · United States | 272k | 104† |
| openai/ | US · United States | 272k | 104† |
| openai/ | US · United States | 272k | 92† |
| openai/ | US · United States | 272k | 92† |
| openai/ | US · United States | 272k | 179† |
| openai/ | US · United States | 272k | 179† |
| openai/ | US · United States | 400k | n/a |
| openai/ | US · United States | 400k | n/a |
| openai/ | US · United States | 272k | n/a |
| openai/ | US · United States | 272k | n/a |
| openai/ | US · United States | 272k | 95† |
| openai/ | US · United States | 272k | 95† |
| openai/ | US · United States | 272k | 77† |
| openai/ | US · United States | 272k | 77† |
| openai/ | US · United States | 272k | n/a |
| openai/ | US · United States | 272k | n/a |
| openai/ | US · United States | 272k | 113† |
| openai/ | US · United States | 1M | 100† |
| openai/ | US · United States | 1M | 100† |
| openai/ | US · United States | 272k | 283† |
| openai/ | US · United States | 272k | 283† |
| openai/ | US · United States | 272k | 172† |
| openai/ | US · United States | 272k | 172† |
| openai/ | US · United States | 1M | n/a |
| openai/ | US · United States | 1M | n/a |
| openai/ | US · United States | 1M | 79† |
| openai/ | US · United States | 1M | 79† |
| openai/ | US · United States | 1M | n/a |
| openai/ | US · United States | 1M | n/a |
| openai/ | US · United States | 922k | 127† |
| openai/ | US · United States | 922k | 131† |
| openai/ | US · United States | 922k | 121† |
| openai/ | US · United States | 922k | 64† |
| openai/ | US · United States | 922k | 124† |
| openai/ | US · United States | 922k | 153† |
| openai/ | US · United States | 922k | 68† |
| openai/ | US · United States | 922k | 96† |
| openai/ | US · United States | 128k | n/a |
| openai/ | US · United States | 128k | n/a |
| openai/ | US · United States | 128k | n/a |
| openai/ | US · United States | 128k | n/a |
| openai/ | US · United States | 128k | n/a |
| openai/ | US · United States | n/a | n/a |
| openai/ | US · United States | n/a | n/a |
| openai/ | US · United States | n/a | n/a |
| openai/ | US · United States | n/a | n/a |
| openai/ | US · United States | n/a | n/a |
| openai/ | US · United States | n/a | n/a |
| openai/ | US · United States | n/a | n/a |
| openai/ | US · United States | n/a | n/a |
| openai/ | US · United States | n/a | n/a |
| openai/ | US · United States | n/a | n/a |
| openai/ | US · United States | n/a | n/a |
| openai/ | US · United States | n/a | n/a |
| openai/ | US · United States | 32k | n/a |
| openai/ | US · United States | 32k | n/a |
| openai/ | US · United States | 128k | n/a |
| openai/ | US · United States | 128k | n/a |
| openai/ | US · United States | 128k | n/a |
| openai/ | US · United States | 32k | n/a |
| openai/ | US · United States | 32k | n/a |
| openai/ | US · United States | 128k | n/a |
| openai/ | US · United States | 16k | n/a |
| openai/ | US · United States | n/a | n/a |
| openai/ | US · United States | n/a | n/a |
| openai/ | US · United States | 200k | 164† |
| openai/ | US · United States | 200k | n/a |
| openai/ | US · United States | 200k | n/a |
| openai/ | US · United States | 200k | n/a |
| openai/ | US · United States | 32k | n/a |
| openai/ | US · United States | 32k | n/a |
| openai/ | US · United States | 8k | n/a |
| openai/ | US · United States | 8k | n/a |
| openai/ | US · United States | 8k | n/a |
| openai/ | US · United States | n/a | n/a |
| scaleway | Region | Context | tok/s |
|---|---|---|---|
| scaleway/ | EU · Paris, France | 8k | n/a |
| scaleway/ | EU · Paris, France | 256k | 154† |
| scaleway/ | EU · Paris, France | 256k | 173† |
| scaleway/ | EU · Paris, France | 256k | 133‡ |
| scaleway/ | EU · Paris, France | 256k | 127‡ |
| scaleway/ | EU · Paris, France | 256k | 140‡ |
| scaleway/ | EU · Paris, France | 128k | 152‡ |
| scaleway/ | EU · Paris, France | 100k | n/a |
| scaleway/ | EU · Paris, France | 180k | 48‡ |
| scaleway/ | EU · Paris, France | 128k | n/a |
| scaleway/ | EU · Paris, France | 250k | 46† |
| scaleway/ | EU · Paris, France | 32k | n/a |
| scaleway/ | EU · Paris, France | 250k | n/a |
| scaleway/ | EU · Paris, France | 256k | n/a |
| scaleway/ | EU · Paris, France | 256k | 45† |
| scaleway/ | EU · Paris, France | n/a | n/a |
| vertex | Region | Context | tok/s |
|---|---|---|---|
| vertex/ | EU · EU multi-region | 1M | n/a |
| vertex/ | EU · EU multi-region | 1M | n/a |
| vertex/ | EU · EU multi-region | 1M | 124† |
| vertex/ | EU · EU multi-region | 1M | 327† |
| vertex/ | EU · EU multi-region | 1M | 203† |
| vertex/ | EU · EU multi-region | 1M | 429† |
| vertex/ | EU · EU multi-region | 1M | 139‡ |
| vertex/ | EU · EU multi-region | 1M | n/a |
| vertex/ | EU · EU multi-region | 1M | 96‡ |
| vertex/ | EU · EU multi-region | 131k | n/a |
| vertex/ | EU · EU multi-region | 1M | 88† |
| vertex/ | EU · EU multi-region | 1M | 94† |
| vertex/ | EU · EU multi-region | 1M | 245† |
| vertex/ | EU · EU multi-region | 2k | n/a |
| vertex/ | EU · EU multi-region | 2k | n/a |
| vertex/ | EU · EU multi-region | 2k | n/a |
| xai | Region | Context | tok/s |
|---|---|---|---|
| xai/ | US · United States | 1M | n/a |
| xai/ | US · United States | 1M | 144† |
| xai/ | US · United States | 1M | n/a |
| xai/ | US · United States | 1M | 92† |
| xai/ | US · United States | 500k | 55† |
| xai/ | US · United States | 500k | 77† |
| xai/ | US · United States | 500k | 73† |
| xai/ | US · United States | 256k | 90† |
tok/s is output throughput, measured from real requests through Sluis over the last 7 days, not from a synthetic benchmark.
- A plain number excludes the wait for the first token.
- A ‡ still includes it.
- A † marks a figure published by the model vendor via Artificial Analysis, which we did not measure.
- n/a means we have no traffic for this model yet and the vendor publishes nothing.
# this key's catalog with this organisation's prices: EUR per million tokens, excluding VAT curl -s https://api.sluis.ai/v1/models \ -H "Authorization: Bearer $SLUIS_KEY" \ | jq '.data[] | {id, input: .pricing.input_eur_per_mtok, output: .pricing.output_eur_per_mtok}' # one model: GET /v1/models/mistral/mistral-large-latest
{
"object": "list",
"data": [
{
"id": "mistral/mistral-large-latest",
"object": "model",
"created": 1704067200,
"owned_by": "mistral",
"kind": "text",
"context_window": 128000,
"capabilities": { "tools": true, "vision": false, "reasoning": null },
"residency": { "jurisdiction": "EU", "region": "Paris, France", "country": "FR", "owner_is_eu": true },
"resolves_to": null,
"pricing": {
"input_eur_per_mtok": 0.55,
"output_eur_per_mtok": 1.65,
"cached_input_eur_per_mtok": 0.055,
"audio_input_eur_per_mtok": null,
"audio_output_eur_per_mtok": null,
"long_context": [{ "from_tokens": 200000, "input_eur_per_mtok": 1.1, "output_eur_per_mtok": 3.3 }],
"eur_per_page": null,
"eur_per_minute": null,
"eur_per_image": null,
"estimated": false
}
},
{
"id": "sluis/fast",
"object": "model",
"created": 1704067200,
"owned_by": "sluis",
"kind": "text",
"context_window": 128000,
"capabilities": { "tools": true, "vision": false, "reasoning": null },
"residency": { "jurisdiction": "EU", "region": "Paris, France", "country": "FR", "owner_is_eu": true },
"resolves_to": "mistral/mistral-large-latest",
"pricing": { "input_eur_per_mtok": 0.55, "output_eur_per_mtok": 1.65, "cached_input_eur_per_mtok": 0.055, "audio_input_eur_per_mtok": null, "audio_output_eur_per_mtok": null, "long_context": [{ "from_tokens": 200000, "input_eur_per_mtok": 1.1, "output_eur_per_mtok": 3.3 }], "eur_per_page": null, "eur_per_minute": null, "eur_per_image": null, "estimated": false }
},
{
"id": "sluis/auto",
"object": "model",
"created": 1704067200,
"owned_by": "sluis",
"kind": null,
"context_window": null,
"capabilities": { "tools": null, "vision": null, "reasoning": null },
"residency": null,
"resolves_to": null,
"pricing": null
},
{
"id": "sluis/ocr",
"object": "model",
"created": 1704067200,
"owned_by": "sluis",
"kind": "ocr",
"context_window": null,
"capabilities": { "tools": null, "vision": null, "reasoning": null },
"residency": null,
"resolves_to": null,
"pricing": {
"input_eur_per_mtok": null,
"output_eur_per_mtok": null,
"cached_input_eur_per_mtok": null,
"audio_input_eur_per_mtok": null,
"audio_output_eur_per_mtok": null,
"long_context": [],
"eur_per_page": 0.01,
"eur_per_minute": null,
"eur_per_image": null,
"estimated": false
}
}
]
}# models callable through Sluis platform credentials, no auth needed curl https://api.sluis.ai/public/models.json | jq '.models[] | "\(.provider)/\(.model)"' # per key: the models your policy, credentials and allow-list actually reach curl https://api.sluis.ai/v1/models -H "Authorization: Bearer $SLUIS_KEY"