wandler

npm install -g wandler or npx wandler --llm <org/repo:precision>

# LLM
wandler --llm onnx-community/gemma-4-E4B-it-ONNX:q4
# LLM with the Wandler serving backend (default)
wandler --backend wandler --llm onnx-community/gemma-4-E4B-it-ONNX:q4 --device webgpu
# Direct transformers.js baseline for A/B testing
wandler --backend transformersjs --llm onnx-community/gemma-4-E4B-it-ONNX:q4 --device webgpu
# LLM on CPU with fp16
wandler --llm LiquidAI/LFM2.5-1.2B-Instruct-ONNX:fp16 --device cpu
# LLM + embeddings
wandler --llm onnx-community/Qwen3.5-0.8B-Text-ONNX:q4 --embedding Xenova/all-MiniLM-L6-v2:q8
# LLM + embeddings + STT
wandler --llm onnx-community/gemma-4-E4B-it-ONNX:q4 --embedding Xenova/all-MiniLM-L6-v2:q8 --stt onnx-community/whisper-tiny:q4
# custom port, auth, listen on all interfaces
wandler --llm LiquidAI/LFM2.5-1.2B-Instruct-ONNX:q4 --port 3000 --host 0.0.0.0 --api-key mysecret