Initial commit

third_party/llamafile/tinyblas_cpu_sgemm_amd_avxvnni.cpp

+// Adapted from
+// https://github.com/Mozilla-Ocho/llamafile/blob/0.8.8/llamafile/tinyblas_cpu_sgemm_amd_avxvnni.cpp
+// Copyrigth 2024 Mozilla Foundation.
+// Copyright(c) 2024 by KVCache.AI, All Rights Reserved.
+
+#ifdef __x86_64__
+#define llamafile_sgemm llamafile_sgemm_amd_avxvnni
+#include "tinyblas_cpu_sgemm.inc"
+#endif  // __x86_64__

third_party/llamafile/tinyblas_cpu_sgemm_amd_fma.cpp

+// Adapted from
+// https://github.com/Mozilla-Ocho/llamafile/blob/0.8.8/llamafile/tinyblas_cpu_sgemm_amd_fma.cpp
+// Copyrigth 2024 Mozilla Foundation.
+// Copyright(c) 2024 by KVCache.AI, All Rights Reserved.
+
+#ifdef __x86_64__
+#define llamafile_sgemm llamafile_sgemm_amd_fma
+#include "tinyblas_cpu_sgemm.inc"
+#endif  // __x86_64__

third_party/llamafile/tinyblas_cpu_sgemm_amd_zen4.cpp

+// Adapted from
+// https://github.com/Mozilla-Ocho/llamafile/blob/0.8.8/llamafile/tinyblas_cpu_sgemm_amd_zen4.cpp
+// Copyrigth 2024 Mozilla Foundation.
+// Copyright(c) 2024 by KVCache.AI, All Rights Reserved.
+
+#ifdef __x86_64__
+#define llamafile_sgemm llamafile_sgemm_amd_zen4
+#define iqk_mul_mat iqk_mul_mat_zen4
+#include "tinyblas_cpu_sgemm.inc"
+#endif  // __x86_64__

third_party/llamafile/tinyblas_cpu_sgemm_arm80.cpp

+// Adapted from
+// https://github.com/Mozilla-Ocho/llamafile/blob/0.8.8/llamafile/tinyblas_cpu_sgemm_arm80.cpp
+// Copyrigth 2024 Mozilla Foundation.
+// Copyright(c) 2024 by KVCache.AI, All Rights Reserved.
+
+#ifdef __aarch64__
+#define llamafile_sgemm llamafile_sgemm_arm80
+#include "tinyblas_cpu_sgemm.inc"
+#endif  // __aarch64__

third_party/llamafile/tinyblas_cpu_sgemm_arm82.cpp

+// Adapted from
+// https://github.com/Mozilla-Ocho/llamafile/blob/0.8.8/llamafile/tinyblas_cpu_sgemm_arm82.cpp
+// Copyrigth 2024 Mozilla Foundation.
+// Copyright(c) 2024 by KVCache.AI, All Rights Reserved.
+
+#ifdef __aarch64__
+#define llamafile_sgemm llamafile_sgemm_arm82
+#define iqk_mul_mat iqk_mul_mat_arm82
+#include "tinyblas_cpu_sgemm.inc"
+#endif  // __aarch64__

third_party/llamafile/tinyblas_cpu_unsupported.cpp

+// Adapted from
+// https://github.com/Mozilla-Ocho/llamafile/blob/0.8.8/llamafile/tinyblas_cpu_unsupported.cpp
+// Copyrigth 2024 Mozilla Foundation.
+// Copyright(c) 2024 by KVCache.AI, All Rights Reserved.
+
+// -*- mode:c++;indent-tabs-mode:nil;c-basic-offset:4;coding:utf-8 -*-
+// vi: set et ft=cpp ts=4 sts=4 sw=4 fenc=utf-8 :vi
+//
+// Copyright 2024 Mozilla Foundation
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+#include "sgemm.h"
+
+bool llamafile_sgemm_unsupported(long m, long n, long k, const void* A, long lda, const void* B, long ldb, void* C, long ldc, int ith, int nth, int task, int Atype, int Btype, int Ctype, int precision) {
+    return false;
+}
+
+bool llamafile_mixmul_unsupported(const struct ggml_compute_params* params,
+                                  const struct ggml_tensor* weights,
+                                  const struct ggml_tensor* thought,
+                                  const struct ggml_tensor* plan,
+                                  struct ggml_tensor* result) {
+    return false;
+}
+
+bool iqk_mul_mat_moe_unsupported(long, long, long, int, int, const void*, const void*, float*, long, long, const void*, int, int) {
+    return false;
+}

pybind11 @ bb05e081

+Subproject commit bb05e0810b87e74709d9f4c4545f1f57a1b386f5