Why the restriction to safetensors and GGUF? onnx is the more common format and specifically made for interchange.
Does anybody use ONNX for LLMs? I've only encountered it for small models packaged to run on consumer devices, and never seen it show up in any workstation or datacenter scenario.
removed
Does anybody use ONNX for LLMs? I've only encountered it for small models packaged to run on consumer devices, and never seen it show up in any workstation or datacenter scenario.