llama.cpp Gains Support for DFM Mimir 1B via HrmText
September 16, 2026
The llama.cpp project has added support for the HrmTextForCausalLM architecture, enabling efficient local inference for the DFM Mimir 1B model. This expands the range of small-scale models available for quantized execution.
HOW THIS AFFECTS YOU
●
builderYou can now run the DFM Mimir 1B model locally using llama.cpp's optimized inference.