BeeAI (gateway)

Route BeeAI’s underlying LLM calls through the Respan gateway to use 250+ models from different providers. Only your RESPAN_API_KEY is needed — no separate provider keys required.

Setup

Install packages

$ pip install respan-ai openinference-instrumentation-beeai beeai-framework

Set environment variables

$ export RESPAN_API_KEY="YOUR_RESPAN_API_KEY"
$ export OPENAI_API_KEY="YOUR_RESPAN_API_KEY"
$ export OPENAI_BASE_URL="https://api.respan.ai/api"

No real OpenAI key needed — the Respan gateway handles provider authentication.

Initialize and run

1 import os
2 
3 os.environ["OPENAI_API_KEY"] = os.environ["RESPAN_API_KEY"]
4 os.environ["OPENAI_BASE_URL"] = "https://api.respan.ai/api"
5 
6 from beeai import Agent
7 
8 agent = Agent(
9     model="gpt-4.1-nano",
10     instructions="You are a helpful research assistant.",
11 )
12 
13 print(agent.run("Tell me about recursion."))

Switch models

Change the model parameter to use 250+ models from different providers through the same gateway.

1 agent = Agent(model="claude-sonnet-4-5-20250929", instructions="...")
2 agent = Agent(model="gemini-2.5-flash", instructions="...")

See the full model list.

$	export RESPAN_API_KEY="YOUR_RESPAN_API_KEY"
$	export OPENAI_API_KEY="YOUR_RESPAN_API_KEY"
$	export OPENAI_BASE_URL="https://api.respan.ai/api"

1	import os
2
3	os.environ["OPENAI_API_KEY"] = os.environ["RESPAN_API_KEY"]
4	os.environ["OPENAI_BASE_URL"] = "https://api.respan.ai/api"
5
6	from beeai import Agent
7
8	agent = Agent(
9	model="gpt-4.1-nano",
10	instructions="You are a helpful research assistant.",
11	)
12
13	print(agent.run("Tell me about recursion."))

1	agent = Agent(model="claude-sonnet-4-5-20250929", instructions="...")
2	agent = Agent(model="gemini-2.5-flash", instructions="...")