Skip to main content
Use SynthClient (sync) or AsyncSynthClient (async) as the only canonical entry point.

​
Sync

from synth_ai import SynthClient

client = SynthClient(api_key="sk_live_...")

system = client.optimization.systems.create(...)
job = client.optimization.offline.create(...)
session = client.optimization.online.create(...)
completion = client.inference.chat.completions.create(
    model="gpt-4o-mini",
    messages=[{"role": "user", "content": "hello"}],
)

​
Async

from synth_ai import AsyncSynthClient

client = AsyncSynthClient(api_key="sk_live_...")
session = await client.optimization.online.create(...)

​
Namespaces

  • client.optimization.systems
  • client.optimization.offline
  • client.optimization.online
  • client.inference.chat.completions
  • client.inference.jobs
  • client.graphs
  • client.verifiers
  • client.pools
  • client.container