Configuration

Squeez resolves backend configuration in order: CLI args > env vars > config file.

Config file

Create a squeez.yaml in your project root, or use configs/default.yaml:

backend: null  # auto-detect from model; or "transformers", "vllm", "encoder"
local_model_path: "./output/squeez_qwen"

# Or remote API backend
# server_url: "https://api.groq.com/openai/v1"
# server_model: "squeez"

Config files are searched in order:

1. ./squeez.yaml
2. ./configs/default.yaml
3. ~/.config/squeez/config.yaml

Environment variables

Variable	Description
SQUEEZ_LOCAL_MODEL	Path to local model directory
SQUEEZ_SERVER_URL	OpenAI-compatible API URL
SQUEEZ_SERVER_MODEL	Remote model ID on that server
SQUEEZ_API_KEY	API key (also checks OPENAI_API_KEY)
SQUEEZ_BACKEND	Optional backend preference: transformers, vllm, or encoder

export SQUEEZ_LOCAL_MODEL=./output/squeez_qwen
# or
export SQUEEZ_SERVER_URL=https://api.groq.com/openai/v1
export SQUEEZ_SERVER_MODEL=squeez
export SQUEEZ_API_KEY=gsk_...

Legacy env vars SQUEEZ_MODEL_PATH and SQUEEZ_BASE_URL still work.

CLI flags

CLI flags override everything:

squeez "Fix the bug" --local-model ./output/squeez_qwen --input-file output.txt
squeez "Fix the bug" --server-url http://localhost:8000/v1 --server-model squeez

Legacy flags --model-path, --base-url, and --model-name still work.

Using with Groq

export SQUEEZ_SERVER_URL=https://api.groq.com/openai/v1
export SQUEEZ_API_KEY=gsk_...
cat output.txt | squeez "Fix the CSRF bug" --server-model openai/gpt-oss-120b