37 tok/s from Qwen3-Next-80B (48 GB) on a 32 GB laptop, 24 without drafts vs 2.8 for llama.cpp; reading 62 vs 16 tok/s. Faster than llama.cpp on all 13 models compared in Q4_K_M. LLMs bigger than your RAM, pure C, CPU or GPU.
C avx2ccpu-inferenceggufinference-enginelinuxllmlocal-llmmixture-of-expertsmoenvmequantizationqwen3simdspeculative-decodingvulkan
/* SPDX-License-Identifier: GPL-3.0-or-later
Copyright (C) 2026 Maurizio Cammalleri */
/*
* deflate.h - the size of a zlib stream (deflate.c).
*/
#ifndef JANAS_COMMON_DEFLATE_H
#define JANAS_COMMON_DEFLATE_H
#include <stddef.h>
#include <stdint.h>
/* the bytes zlib's compress() makes of n bytes at its default level, to
the byte; -1 when out of memory or longer than one window (32 KiB less
262 bytes) */
long janas_deflate_size(const uint8_t *p, size_t n);
#endif