Reasoning and programming benchmarks for small models
GPT-X3-150M on ONNX Runtime, FP32 down to INT4
Chat with our most capable model