Use apimodels in Cherry Studio

Get connected in 3 steps

1. 1

   Obtenha uma chave API

   Abra Console no seu console e crie uma chave sk_….

2. 2

   Add an OpenAI provider

   Cherry Studio → Settings → Model Providers → Add, pick type OpenAI. API host + key below.

3. 3

   Pick a model

   Add models with + or fetch the list. Pick gpt-6.1-sol to chat, gpt-image-2 to generate images — both work in an ordinary conversation, no separate setup.

Provider settings

Common model IDs

The full list shows up in Cherry Studio's model dropdown (served by our /v1/models endpoint). The common ones:

Generate images inside a chat

Image models sit behind the same provider and the same API host as the chat models — no second provider, and no # suffix tricks on the URL. In an ordinary conversation switch the model to gpt-image-2 (or gemini-3.1-flash-image-preview, grok-imagine-image, …) and just send your description; the picture comes back as an image in the reply.

curl -s https://api.apimodels.app/v1/chat/completions \
  -H "Authorization: Bearer $APIMODELS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2",
    "messages": [{ "role": "user", "content": "a cute corgi puppy on grass" }]
  }'
# → assistant message containing: ![a cute corgi puppy on grass](https://r2.apimodels.app/images/….png)

Attach an image to the same message and it becomes image-to-image editing — just describe the change. Note that each message stands alone: the previous picture is not silently reused as a reference, so "draw a cat" followed by "draw a dog" will not bleed together. To edit the previous image, attach it again.

For programmatic use — your own size, count and callbacks — the standard OpenAI image endpoint is still there:

curl -s https://api.apimodels.app/v1/images/generations \
  -H "Authorization: Bearer $APIMODELS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2",
    "prompt": "a cute corgi puppy on grass",
    "n": 1,
    "size": "1024x1024"
  }'

Verificação rápida de 30 segundos

If it won't connect, confirm the key + endpoint with curl first — this is exactly what Cherry Studio hits on "check connection":

# confirm the endpoint + your key work (this is what Cherry Studio checks):
curl -s https://api.apimodels.app/v1/models \
  -H "Authorization: Bearer $APIMODELS_API_KEY" | head -c 300

Then a chat call:

curl -s https://api.apimodels.app/v1/chat/completions \
  -H "Authorization: Bearer $APIMODELS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-6.1-sol",
    "messages": [{ "role": "user", "content": "Reply with exactly: ok" }]
  }'

Solução de problemas