We recently released Smaug-72B-v0.1 which has taken first place on the Open LLM Leaderboard by HuggingFace. It is the first open-source model to have an average score more than 80.
We recently released Smaug-72B-v0.1 which has taken first place on the Open LLM Leaderboard by HuggingFace. It is the first open-source model to have an average score more than 80.
And at 72 billion parameters it’s something you can run on a beefy but not special-purpose graphics card.
Based on the other comments, it seems like this needs 4x as much ram than any consumer card has
It hasn’t been quantized, then. I’ve run 70B models on my consumer graphics card at a reasonably good tokens-per-second rate.
I’m curious how local generation goes with potentially dedicated AI extensions using stuff like tensor cores and their own memory instead of hijacking parts of consumer GPUs for this.