master git branch
llama.cpp output:
D:\cpp\llama.cpp\cubetitled-ui\llama.cpp\ggml\src\ggml-backend-meta.cpp:625: shape mismatch for VIEW
I use these commands to start:
set LL_MODEL=Qwen_Qwen3.6-27B-Q5_K_L
:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
set LLM_TEMP=--temp 0.6 --top-k 20
set CTX_SIZE_OPTS= --ctx-size 10000
set VRAM_LOAD_OPTS= --fit-target 100 --n-gpu-layers 999
set REP_PENALITY_OPTS= --repeat-penalty 1.01 --repeat-last-n 8192
set MODEL_OPTS= -m models\%LL_MODEL%\%LL_MODEL%.gguf --alias "%LL_MODEL%"
set CHAT_TEMPL_OPTS= --jinja --chat-template-file models\%LL_MODEL%\chat_template.jinja
set MISC_OPTS= --port 11112 --log-colors off --no-mmap --parallel 1 -t 12
set THINK_OPTS= --reasoning-preserve --chat-template-kwargs "{\"preserve_thinking\":true}"
set GPU_OPTS= --split-mode tensor
set CACHE_PROMPT_OPTS= --cache-prompt --cache-ram 16384 --cache-reuse 8 -sps 0.8
:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
title %LL_MODEL% `%LLM_TEMP%` + cubetitled-ui
set path=D:\cpp\llama.cpp\cubetitled-ui\llama.cpp\build\bin\Release;%path%
llama-server.exe %MODEL_OPTS% %CHAT_TEMPL_OPTS% %VRAM_LOAD_OPTS% %CTX_SIZE_OPTS% %LLM_TEMP% %MISC_OPTS% %MTP_OPTS% %THINK_OPTS% %GPU_OPTS% %REP_PENALITY_OPTS% %CACHE_PROMPT_OPTS%
mastergit branchllama.cpp output:
I use these commands to start: