One key for many models.
FlowModels is an OpenAI-compatible API. You name the model in the model field, pay from a euro balance and see what each model costs before you call it.
In the catalog today: 77 models from 12 providers.
For businesses and self-employed professionals only
import os
from openai import OpenAI
client = OpenAI(
base_url="https://api.flowmodels.one/v1",
api_key=os.environ["FLOWMODELS_API_KEY"],
)
response = client.chat.completions.create(
model="anthropic/claude-haiku-5-5",
messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)
Pick a model on the right: only the model line changes, your key and billing stay the same.
The lowest-priced chat model per provider
Euros per 1M tokensAs of Nov 22, 2026 All models and prices
Moving from the OpenAI library: two lines and a model name.
The API speaks the OpenAI format. You enter the base URL and your key, and the rest of your code stays as it is. From then on, you name the model as provider/model.
client = OpenAI(
- base_url="https://api.openai.com/v1",
+ base_url="https://api.flowmodels.one/v1",
- api_key=os.environ["OPENAI_API_KEY"],
+ api_key=os.environ["FLOWMODELS_API_KEY"],
)
What you can call today.
-
POST /v1/chat/completionsChat responses, with or without streaming -
POST /v1/responsesResponses with stored history and background runs -
POST /v1/embeddingsVectors for search and matching -
POST /v1/rerankOrder documents by relevance to a query, hosted in the EU -
POST /v1/audio/transcriptionsTurn audio files into text -
POST /v1/videosVideos from text or a start image, as a background job -
POST /v1/images/generationsImages from text, right in the response -
GET /v1/modelsEvery callable model with its prices, machine-readable
What a call cost is right there in the log.
Every request appears in your dashboard's request log with its model, provider, tokens and cost. You top up credit from €20; there is no starting credit and no subscription.