One API key.
Every model.
Prism gives you a single API key and a single bill for OpenAI and Gemini models. Fully compatible with the OpenAI SDK — the only change required is the base URL and API key.
from openai import OpenAI
client = OpenAI(
base_url="https://your-prism-domain.com/api/v1", # only change
api_key="sk-dprism-...", # only change
)
completion = client.chat.completions.create(
model="openai/gpt-5.4-mini",
messages=[{"role": "user", "content": "Hello!"}],
)
print(completion.choices[0].message.content)Your existing OpenAI SDK code works unchanged — same methods, same streaming, same response shape.
Fully compatible with OpenAI SDK
Use the OpenAI Python/JS SDK you already know, with just a base URL change. Chat, streaming, and image generation all work as-is.
One key, every model
OpenAI and Gemini behind a single API key and a single token balance. Switch providers by changing the model string.
Transparent token billing
Every call is billed on the model's real usage — see exact tokens used per call in your logs.
Available models
Starting with a curated set of fast, low-cost models from OpenAI and Google.
GPT-5.4 Nanoopenaichat
openai/gpt-5.4-nano26.67 tokens / 1M input tokens
166.67 tokens / 1M output tokens
166.67 tokens / 1M output tokens
GPT-5.4 Miniopenaichat
openai/gpt-5.4-mini100.00 tokens / 1M input tokens
600.00 tokens / 1M output tokens
600.00 tokens / 1M output tokens
Gemini Flash-Litegeminichat
gemini/gemini-flash-lite-latest33.33 tokens / 1M input tokens
200.00 tokens / 1M output tokens
200.00 tokens / 1M output tokens
Gemini Flashgeminichat
gemini/gemini-flash-latest200.00 tokens / 1M input tokens
1200.00 tokens / 1M output tokens
1200.00 tokens / 1M output tokens
Gemini Flash-Lite Imagegeminiimage
gemini/gemini-3.1-flash-lite-image33.33 tokens / 1M input tokens
4000.00 tokens / 1M output tokens
4000.00 tokens / 1M output tokens
GPT Image 1 Miniopenaiimage
openai/gpt-image-1-mini266.67 tokens / 1M input tokens
1066.67 tokens / 1M output tokens
1066.67 tokens / 1M output tokens