Retrieve a model

GET/models/{model}

Returns one entry of the catalogue — what models.retrieve() calls in the OpenAI SDKs — with the model's full compliance record. No key needed; an ID Evanesca does not serve answers 404 model_not_found. Embedding and image entries carry operator, location, retention and training too.

Catalogue checked on 20 Oct 2026

SoonOpening soon: what follows describes the service as it will work.

Server URLhttps://evanesca.app/api/v1

GET/models/{model}Opens at launch

Authorizationnone — no key needed

Path parameters

modelstringrequired
A model ID, as GET /models lists it — for example gemma-4-26b.

Fields of an entry

application/json

idstring
The model ID: the value of model in your calls.
objectstring
model, as in OpenAI's format.
kindstring
The model's family.
Valueschatembeddingimage
createdinteger
Unix time, as in OpenAI's format.
owned_bystring
The model's maker.
operatorstring
Who runs the model — the value of the x-evanesca-operator header.
Valuesmistralakiscalewayovhcloudinfomaniak
locationstring
Where it runs — the value of the x-evanesca-location header.
Valuesparisgravelinesmistral-eu-regiongermanyswitzerland
context_lengthinteger
The context window as the operator announces it; absent while it has not been checked.
tool_callingboolean
true or false where tool calling was measured with a real call; absent where it was not.
reasoningboolean
Whether the model thinks before it answers.
pricingobject
currency, vat_included, input_per_million and output_per_million: euros per million tokens, VAT included.
licence, open_weightsstring | null · boolean
The licence as published — null while none is — and whether the weights are public.
checksobject
The six checks as booleans — licence, provenance, security, gdpr, benchmarks, live_test — and, in run_on, the date of each passed check where it is known. false means missing: not run yet, or run without meeting the pass mark.
retentionobject
What the operator says it keeps (rule), and the page where it says so (source). Evanesca itself stores no prompt.
trainingobject
What the operator says about training on your data (statement), and its source (source).

Where it runs, what it costs

Evanesca · on the record

Answered by: Evanesca's catalogue: no model runs, no operator is called.
Places listed: paris (Paris, France), gravelines (Gravelines, France), mistral-eu-region (EU region of Mistral AI), germany (Germany) and switzerland (Switzerland)
Costs: Nothing: no key, no credit.

Errors

Status	type	Meaning
404	`model_not_found`	The ID is unknown, or a model Evanesca does not serve.

Other operator errors reach you with its status and body, and Retry-After when it sends one (errors).

curl https://evanesca.app/api/v1/models/gemma-4-26b

model = client.models.retrieve("gemma-4-26b")
print(model.operator, model.location)

const model = await client.models.retrieve("gemma-4-26b");
console.log(model.operator, model.location);

package main

import (
	"fmt"
	"io"
	"net/http"
)

func main() {
	req, err := http.NewRequest("GET", "https://evanesca.app/api/v1/models/gemma-4-26b", nil)
	if err != nil {
		panic(err)
	}
	res, err := http.DefaultClient.Do(req)
	if err != nil {
		panic(err)
	}
	defer res.Body.Close()
	out, _ := io.ReadAll(res.Body)
	fmt.Println(res.Status, string(out))
}

{
  "id": "gemma-4-26b",
  "object": "model",
  "kind": "chat",
  "created": 1792454400,
  "owned_by": "Google",
  "operator": "aki",
  "location": "germany",
  "context_length": 256000,
  "tool_calling": true,
  "reasoning": false,
  "pricing": {
    "currency": "EUR",
    "vat_included": true,
    "input_per_million": 0.168,
    "output_per_million": 0.84
  },
  "licence": "Apache-2.0",
  "open_weights": true,
  "checks": {
    "licence": true,
    "provenance": true,
    "security": true,
    "gdpr": true,
    "benchmarks": true,
    "live_test": true,
    "run_on": {
      "security": "2026-10-13",
      "benchmarks": "2026-10-13",
      "live_test": "2026-10-13"
    }
  },
  "retention": {
    "rule": "Inference content “is processed only in volatile memory […] and is not logged, stored, or analyzed by AKI.IO”; usage data (request ID, model endpoint, status, timestamps, token usage) “is retained for 90 days”.",
    "source": "https://aki.io/privacy-policy"
  },
  "training": {
    "statement": "“Inference requests are not forwarded to model developers, and customer prompts and outputs are not used to train AI models.”",
    "source": "https://aki.io/"
  },
  "status": "live"
}

{
  "error": {
    "message": "Unknown model \"my-model\". …",
    "type": "model_not_found"
  }
}