In one of your interview, you was able to run and measure performance of Llama2-7B LLM at up to 10 tokens per second. Can you share HEF, or, at least, HAR file for us, for exploratory purposes? If the answer is “no”, maybe you can share the HEF or HAR with zeroed parameters? Thank you.
The Hailo-10 is currently not yet available for a wider audience. We need to ask your for patience until we can release LLM models for you to try.