ownapibridge 0.1.0

Python client for the ownAPIbridge API — access 100+ AI models with one key

ownAPIbridge Python SDK

Access 10+ AI models (GPT, Claude, Gemini, DALL·E, and more) through a single API key.

Install

pip install ownapibridge
# or from source:
pip install -e sdk/python

Quick start

from ownapibridge import OABClient

client = OABClient(
    api_key="sk-oab-your-key-here",
    base_url="http://localhost:8000",   # or your deployed server
)

# --- Chat ---
response = client.chat.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "What is the capital of France?"}],
)
print(response.content)   # Paris
print(response.total_tokens)

# --- Streaming ---
for chunk in client.chat.stream(model="claude-sonnet-4", messages=[
    {"role": "user", "content": "Write a haiku about code."}
]):
    print(chunk, end="", flush=True)

# --- List models ---
models = client.models.list()
for m in models:
    print(m.id, m.provider, f"${m.price_per_1m_input}/1M in")

# --- Image generation ---
image = client.images.generate(
    model="dall-e-3",
    prompt="A red fox in the snow, photorealistic",
)
print(image.url)

# --- Video generation ---
video = client.videos.generate(
    model="kling-v1-5",
    prompt="A timelapse of a flower blooming",
    duration=5,
)
print(video.url)

Error handling

from ownapibridge import OABClient, AuthenticationError, ModelNotFoundError, ProviderError

try:
    response = client.chat.create(model="gpt-4o", messages=[...])
except AuthenticationError:
    print("Invalid or expired API key")
except ModelNotFoundError:
    print("Model not found")
except ProviderError as e:
    print(f"Upstream provider error: {e}")

Zero dependencies

The SDK uses only Python's standard library (urllib, json, dataclasses). No pip installs required beyond the package itself.

API reference

OABClient(api_key, base_url)

Resource	Methods
client.chat	.create(model, messages, temperature, max_tokens, system) → ChatResponse
client.chat	.stream(model, messages, ...) → Iterator[str]
client.models	.list() → list[ModelInfo]
client.images	.generate(model, prompt, size, quality, n) → ImageResponse
client.videos	.generate(model, prompt, duration, aspect_ratio) → VideoResponse

Available models [id (name) - provider])

• gemini-3-1-pro (Gemini 3.1 Pro Preview) - google
• gpt-5.4 (GPT-5.4) - openai
• claude-opus-4 (Claude Opus 4.6) - anthropic
• claude-sonnet-4 (Claude Sonnet 4.6) - anthropic
• gpt-4o (GPT-4o) - openai
• gemini-3-flash (Gemini 3 Flash) - google
• claude-haiku-4 (Claude Haiku 4.5) - anthropic
• gpt-4o-mini (GPT-4o Mini) - openai
• deepseek-v3 (DeepSeek V3.2) - deepseek
• o3 (o3) - openai
• o4-mini (o4-mini) - openai
• gpt-4-1 (GPT-4.1) - openai
• gpt-4-1-mini (GPT-4.1 Mini) - openai
• gemini-2-0-flash (Gemini 2.0 Flash) - google
• llama-4-maverick (Llama 4 Maverick) - groq
• llama-3-3-70b (Llama 3.3 70B) - groq
• qwq-32b (QwQ-32B) - groq
• deepseek-r1 (DeepSeek R1) - deepseek