mikhail/llama.cpp-mtp-turboquant
1
0
Fork 0
Code Issues Pull Requests Actions 29 Packages Projects Releases Wiki Activity

llama: end-to-end tests (#19802)

Browse Source 
* tests: add end-to-end tests per model architecture

* fixup for rebase

* fix use-after-free in llama-model-loader.cpp

* fix CI

* fix WebGPU

* fix CI

* disable CI for macOS-latest-cmake-arm64

* use expert_weights_scale only if != 0.0f

* comments
This commit is contained in:
Johannes Gäßler
2026-03-08 12:30:21 +01:00
committed by GitHub
parent a95047979a
commit a976ff081b
33 changed files with 1607 additions and 633 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/llama-model-saver.h
+5 -2
View File
@@ -1,5 +1,6 @@
#pragma once
#include "gguf.h"
#include "llama.h"
#include "llama-arch.h"
@@ -7,10 +8,12 @@
struct llama_model_saver {
struct gguf_context * gguf_ctx = nullptr;
const struct llama_model & model;
const bool gguf_ctx_owned;
const struct llama_model * model;
const struct LLM_KV llm_kv;
llama_model_saver(const struct llama_model & model);
llama_model_saver(const struct llama_model * model);
llama_model_saver(enum llm_arch arch, struct gguf_context * gguf_ctx);
~llama_model_saver();
void add_kv(enum llm_kv key, uint32_t value);