How to Connect Open Interpreter to APIVAI (Cheap Claude & GPT)

Connect Open Interpreter to APIVAI

Open Interpreter lets an LLM run code on your
machine to complete tasks. It uses LiteLLM under the hood, so any OpenAI-compatible endpoint works —
point it at APIVAI to run on Claude or GPT at a fraction of list price.

Configure

Set the model, API base, and key (env vars or flags):

export OPENAI_API_KEY="YOUR_APIVAI_API_KEY"
export OPENAI_API_BASE="https://api.apivai.com/v1"
interpreter --model openai/gpt-5.5

Or in Python:

from interpreter import interpreter
interpreter.llm.model = "openai/gpt-5.5"
interpreter.llm.api_base = "https://api.apivai.com/v1"
interpreter.llm.api_key = "YOUR_APIVAI_API_KEY"
interpreter.chat("Plot the CSV in ./data and save a PNG")

The openai/ prefix tells LiteLLM to use the OpenAI-compatible path.

Pick a model

curl -s https://api.apivai.com/v1/models -H "Authorization: Bearer $APIVAI_API_KEY"

Use a returned name (Claude Sonnet is a strong default for code/agent tasks).

Tips

• Agent runs send many tokens — a discounted gateway noticeably cuts the bill.
• Always review/confirm code before it runs; Open Interpreter can execute on your system.
• Pick a large-context model for big tasks.

FAQ

Does Open Interpreter work with APIVAI? Yes — set api_base to https://api.apivai.com/v1,
the openai/<model> model name, and your APIVAI key.

Which model is best? Claude Sonnet for code and multi-step tasks; GPT-5.5 for faster, cheaper runs.

Why use a gateway? Agentic runs burn tokens; APIVAI's discounted OpenAI-compatible access cuts cost.

Get started

Set the api_base + key and run Open Interpreter on a model from /v1/models. Examples:
APIVAI examples repo.