Submissions from twitter.com/artificialanlys

		Grok 4 is now the leading AI model ( ArtificialAnlys) (twitter.com/artificialanlys)
		13 points by JnBrymn 6 months ago \| past \| 6 comments
		DeepSeek V3 is now the highest scoring non-reasoning model (twitter.com/artificialanlys)
		14 points by aurareturn 10 months ago \| past \| 3 comments
		We've now partially replicated Reflection Llama 3.1 70B's eval claims (twitter.com/artificialanlys)
		4 points by _micah_h on Sept 8, 2024 \| past \| 1 comment
		Cerebras launches inference for Llama 3.1; benchmarked at 1846 tokens/s on 8B (twitter.com/artificialanlys)
		95 points by _micah_h on Aug 27, 2024 \| past \| 42 comments
		Sambanova breaks 1000 tokens/SEC on LLama3 8B (twitter.com/artificialanlys)
		7 points by germanjoey on May 28, 2024 \| past
		From GPT-4 to Mistral 7B, there is a 300x range in the cost of LLM inference (twitter.com/artificialanlys)
		2 points by Gcam on Feb 12, 2024 \| past
		Mistral API reduces time to first token by 10x (only place for Mistral Medium) (twitter.com/artificialanlys)
		4 points by Gcam on Feb 5, 2024 \| past
		240 Tokens/s achieved by Groq's custom chips on Lama 2 Chat (70B) (twitter.com/artificialanlys)
		5 points by Gcam on Jan 31, 2024 \| past
		New GPT-4 Turbo (0125 Preview) slightly faster per initial benchmarks (twitter.com/artificialanlys)
		2 points by Gcam on Jan 26, 2024 \| past