
Any updates on running Qwen3.8 27B on R9700 ?
Hi,
I checked this blog post which mentions ~50 tokens/s for Qwen3.8 27B on single R9700.
I checked huggingface but no specific instructions for running it on RCom.
Is there any specific documentation for running it using llama.cpp or vllm ? All the craze is only for Nvidia GPUs :-(