AceInstruct Llama.cpp

The AceInstruct family of LLMs (1.5B, 7B, and 72B parameters) excels in coding, math, and general tasks, matching Qwen2.5-Instruct's performance. Built upon the Qwen architecture and fine-tuned with instruction-following datasets, AceInstruct surpasses the specialized AceMath-Instruct in versatility.

Model

Select the AI model to use for chat

512 2048
0.1 2
0.1 1
1 100
1 2