A simple Rust implementation of a language model using the Burn framework.

Build and run with default settings:

cargo run --release

This will generate text from the prompt "what is the meaning of life?" and produce up to 100 tokens.

The project supports different backends via feature flags:

• CPU backend (default): cargo run --release --features cpu
• Metal backend (macOS): cargo run --release --features metal

The Metal backend provides better performance on Apple Silicon Macs.

Use a different prompt:

cargo run --features cpu --release -- --prompt "Write a short story"

Generate more tokens:

cargo run --features cpu --release -- --max-tokens 200

• --prompt or -p: The text prompt to generate from (default: "what is the meaning of life?")
• --max-tokens or -m: Maximum number of tokens to generate (default: 100)
• --model or -m: The model to use (default: qwen2-0-5b)
  • qwen2-0-5b: 0.5 billion parameter model
  • qwen2-1-5b: 1.5 billion parameter model