Local 1M Context Inference at 15 tokens/s and ~100% "Needle In a Haystack"(old.reddit.com)
old.reddit.com
Local 1M Context Inference at 15 tokens/s and ~100% "Needle In a Haystack"
https://old.reddit.com/r/LocalLLaMA/comments/1f3xfnk/local_1m_context_inference_at_15_tokenss_and_100/
0 comments
—