10 trial calls, up to $1 free credit, no card needed
from openai import OpenAI
client = OpenAI(
base_url="https://synthorai.io/v1",
api_key="sk-syn-…",
)
resp = client.images.generate(
model="seedream-5-0-260128",
prompt="an AI gateway routing data through a neural network, soft lavender light, bright airy studio",
size="2048x2048",
)
seedream-5-0-260128 · 2048² · $0.035 POST /v1/images/generations Four pillars, each with concrete technical backing.
Chat, image, and audio models: OpenAI, Anthropic, Google, AWS Bedrock, Alibaba Qwen, DeepSeek, BytePlus, and more, all behind one OpenAI- or Anthropic-compatible endpoint. Per-request failover and multi-key load balancing keep your app up when an upstream goes down. No code changes beyond baseURL.
Zero retention by default: prompts and completions are never stored. TLS 1.3 in transit, isolated per workspace, with guardrails that scan for PII and secrets before your data ever leaves your perimeter.
Pay less than the upstream's list price. No platform markup, no per-seat fees. Native prompt caching across Anthropic, OpenAI, and Gemini cuts repeat input to ~10% of list. BYOK free and Stripe fees absorbed. Hard quota caps per workspace and per API key keep a runaway script from outrunning your budget.
GDPR and CCPA ready, with a DPA your legal team can actually sign. Pin requests to a region (US / EU / APAC) per workspace, enforce per-workspace quotas, and inspect per-request audit logs for every route, provider, and retention decision. SSO (SAML / OIDC) and fine-grained RBAC included.
Same key, same OpenAI-compatible surface. The call shape follows the modality: chat, image, speech, video, and realtime voice.
Every major LLM behind one endpoint, with prompt caching and hard spend caps.
/v1/chat/completions Text-to-image at list price, billed per image, zero retention by default.
/v1/images/generations Transcribe audio with Whisper or gpt-4o, priced by the minute.
/v1/audio/transcriptions Text- and image-to-video with a create-and-poll job API: Sora, Seedance, and Veo.
/v1/videos Low-latency speech-to-speech over WebSocket, one dialect for gpt-realtime and Gemini Live.
/v1/realtime Under a minute from sign-up to first call. Payment processing fee on us. No sales call to scale.
Google, GitHub, or email. You're in the console in seconds, no invite needed.
BYOK is free with zero platform markup on the upstream list price. Or fund with USD and set hard caps per workspace and per API key.
One OpenAI- or Anthropic-compatible base URL. Paste your key. Your existing code keeps working.