ADOLAMODEL API / ISSUE 001

Ship open models. without rebuilding your stack.

Run GPT-OSS, Qwen, Llama, and DeepSeek through one OpenAI-compatible API. Stream responses, switch models with one parameter, and see exactly what each request costs.

Create your Adola account Try the interactive demo

Usage-based pricing. No recurring subscription, seat fee, or monthly minimum spend.

MODEL SELECTORREQUEST / 0248

01GPT-OSS 120B02Qwen3 30B03Llama 3.3 70B04DeepSeek V4 Flash

ADOLA/v1

200 OKdeepseek-v4-flashSTREAMING

INPUT128OUTPUT74REQUEST COST$0.001

OpenAI-compatibleExact model selectionStreaming responsesPer-request usage

01 / THE CONTRACT

Same client.
4 real choices.
One clear receipt.

Keep the request shape your application already understands. Choose the model in the request, stream the answer, and read standard token usage in the response.

01 Keep your OpenAI client02 Set the Adola base URL03 Name the model you want04 Receive the familiar response

REQUESTPOST /v1/chat/completionsMODELdeepseek-v4-flashRESPONSEchat.completion.chunk

02 / MODEL INDEX

Choose by the work.
Pay by the model.

Use a public model ID in every request. The model you select is the model that responds.

Compare model pricing

GPT-OSS 120B

gpt-oss-120b

OpenAI open-weight 120B MoE for capable reasoning, coding, and general-purpose work.

CONTEXT131KINPUT / 1M$0.14OUTPUT / 1M$0.57

Qwen3 30B

qwen3-30b

Fast multilingual MoE with strong tool use and agentic performance.

CONTEXT41KINPUT / 1M$0.11OUTPUT / 1M$0.47

Llama 3.3 70B

llama-3.3-70b

Meta's proven 70B instruction model for dependable general-purpose workloads.

CONTEXT131KINPUT / 1M$0.56OUTPUT / 1M$0.75

DeepSeek V4 Flash

deepseek-v4-flash

Fast, efficient reasoning and coding with a one-million-token context window.

CONTEXT1MINPUT / 1M$0.13OUTPUT / 1M$0.26

03 / TWO CHANGES

Keep the client.
Change the coordinates.

Point your existing OpenAI client at Adola, use your Adola API key, and choose the model for the work in front of you.

Copy the quickstart

PYTHON / QUICKSTART2 CHANGES

from openai import OpenAI

client = OpenAI(
  base_url="https://api.adola.app/v1",
  api_key="adola_...",
)

stream = client.chat.completions.create(
  model="gpt-oss-120b",
  messages=[...],
  stream=True,
)

response streamingchat.completion.chunk

04 / MAKE THE CALL

Your next model is
one parameter away.

Add a payment method once, then pay only for successfully metered usage at the selected model's published rate.

Create your Adola account See model pricing