Lemonade Runner¶
Lemonade is kew's own tool-use loop (kew-lemonade) over any OpenAI-compatible local endpoint. It supports AMD Lemonade, LM Studio, llama.cpp's llama-server, vLLM, Ollama, and any other OpenAI-compatible server. Cost is $0 (local); nothing leaves the machine.
Prerequisites¶
- Install a local OpenAI-compatible server — see Lemonade's installation guide.
- Load a model into your server.
Selecting a model¶
Set model = "lemonade/<model-id>" in kew.toml; the lemonade/ prefix routes to this runner.
Configuration¶
The [agent.lemonade] block accepts the following settings:
base_url(defaulthttp://localhost:13305/v1) — the OpenAI-compatible endpoint to connect to.max_steps(default10) — the maximum number of tool calls allowed in a single run.
Examples¶
A kew.toml snippet:
[agent]
model = "lemonade/llama3"
[agent.lemonade]
base_url = "http://localhost:13305/v1"
max_steps = 15
Direct invocation with kew-lemonade:
kew-lemonade --base-url http://localhost:13305/v1 --model llama3 -p "Write a Python script to calculate Fibonacci numbers"
Limitations (v1)¶
- Model quality is the model's — kew does not influence model quality; it's entirely up to the model you choose.
- Tool-calling reliability varies by model — some models may struggle with consistent tool calling.
run_bashruns with the loop's privileges — since this runner executes locally,run_bashcommands run with the same privileges as the kew process, so it's recommended to run kew inside a sandbox.