AI正在发生 未来正在更新Llama-3.2-1B-Instruct-NEO-SI-FI-GGUF
It is the new "Llama-3.2-1B-Instruct", max context of 131,000 (128k) with the NEO IMATRIX Science Fiction and Story dataset. The power in this 1B (for its size) is frankly jaw dropping... and at 190 tokens per second + on a GPU. This is a LLAMA 3.2 model, and requires Llama3 template, but may work with other template(s) and has maximum context of 128k.
译文尚未就绪，显示来源原文
It is the new "Llama-3.2-1B-Instruct", max context of 131,000 (128k) with the NEO IMATRIX Science Fiction and Story dataset. The power in this 1B (for its size) is frankly jaw dropping... and at 190 tokens per second + on a GPU. This is a LLAMA 3.2 model, and requires Llama3 template, but may work with other template(s) and has maximum context of 128k.