Run frontier MoE models on hardware you already own — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦
C
#ifndef COLIBRI_NATIVE_QUANT_DUAL_H
#define COLIBRI_NATIVE_QUANT_DUAL_H
#include "tensor.h"
int coli_fp4_dual_matvec_ref(float *output_a, float *output_b,
const ColiTensorView *weight_a,
const ColiTensorView *weight_b,
const float *input);
int coli_fp8_dual_matvec_ref(float *output_a, float *output_b,
const ColiTensorView *weight_a,
const ColiTensorView *weight_b,
const float *input);
#endif