3 articles found
Tencent’s Youtu-LLM-2B challenges LLM scaling laws with 128K context and superior agentic capabilities despite having only 1.96B parameters.
Moonshot AI’s hybrid architecture delivers 6x decoding speed with 75% less memory, making 1M-token contexts actually practical.
Samsung’s Tiny Recursive Model with microscopic 7M parameters beats massive LLMs on reasoning tasks, challenging the ‘bigger is better’ dogma.
