Goose is an open-source AI agent that goes beyond code suggestions: it installs dependencies, runs and tests code, and automates whole engineering workflows from the terminal or its desktop app. Created by Block and now part of the Linux Foundation's Agentic AI Foundation alongside MCP, it is built to be model-agnostic. Opper ships in its provider list, so you pick Opper, paste your key, and Goose runs on any of 700+ models with a live model list.
GooseoninIn Goose (v1.49 or later), run goose configure → Configure Providers → Opper, or in the desktop app open Settings → Models → Configure providers → Opper, then paste your Opper key (stored as OPPER_API_KEY). Then pick Kev 4B.
Kev 4B runs on the routes below. Route only to EU providers when your data has to stay in Europe.
| Provider | Region | Zero data retention | Training | Input | Output | Cache read |
|---|---|---|---|---|---|---|
| EU | No | $0.04 | $0 | Input price |
Kev 4B, Jared Palmer's open decision model fine-tuned from Alibaba's Qwen3.5-4B, answers yes/no, choice and score questions about text with calibrated probabilities instead of writing text.
Step-by-step setup for running Goose on each of these models through Opper.
Anthropic's terminal coding agent. Run it on any model on Opper, not just Claude.
opper launch claude
OpenAI's open-source coding CLI. Run it on any model on Opper.
opper launch codex
DeepSeek's open-source coding agent. Add Opper as a custom provider to run it on any model, not just DeepSeek's.
OpenAI-compatible