IBM Granite 4.0 Nano released: 1.5B - 350M parameter models

AI Impact Summary

IBM has released Granite 4.0 Nano, a family of small language models ranging from 1.5B to 350M parameters, designed for edge and on-device applications. These models leverage IBM's hybrid architecture and training methodologies, achieving competitive performance compared to larger models like Qwen, Gemma, and others, particularly in agentic workflows and key benchmarks. The release highlights IBM's ongoing investment in efficient, high-performing models suitable for resource-constrained environments.