AMD · AMD-Llama
AMD-Llama-135M
AMD's 135M text model, released with training code and used as a speculative-decoding draft model on Ryzen AI hardware.
- Parameters
- 135M
- Input
- Text
- Output
- Text
- Access
- Open source
- Licence
- Apache 2.0
- Released
- Device evidence
- Measured on device
- Last verified
- Formats
- Safetensors
- Runtimes
- Transformers
- Platforms
- Linux, Windows
AMD also publishes a code-oriented variant trained to draft tokens for CodeLlama-7B. Its Ryzen AI figures measure the complete speculative-decoding setup, not the 135M model as a standalone assistant.