thek0tyara/llama-cpp-turboquant
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 4

benches : update models + numbers (#19359)

Browse source 
* bench : update script

* benches : update numbers
This commit is contained in:
Georgi Gerganov 2026-02-05 14:34:07 +02:00 • committed by GitHub
parent b828e18c75
commit 3795cc1e89
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
3 changed files with 541 additions and 188 deletions
Download patch file Download diff file Expand all files Collapse all files

60
scripts/bench-models.sh Normal file → Executable file
View file

@ -7,47 +7,54 @@ ARGS_BB="-c 270336 -npp 512,4096,8192 -npl 1,2,4,8,16,32 -ntg 32"
ARGS_B="-d 0,4096,8192,16384,32768 -p 2048 -n 32"
QUICK=0
DIO=0
while (( "$#" )); do
case "$1" in
--quick) QUICK=1; shift ;;
*) shift ;;
esac
case "$1" in
--quick) QUICK=1; shift ;;
--dio) DIO=1; shift ;;
*) shift ;;
esac
done
if (( QUICK )); then
ARGS_BB="-c 20480 -npp 512,4096 -npl 1,2,4 -ntg 32"
ARGS_B="-d 0 -p 2048 -n 32"
ARGS_BB="-c 20480 -npp 512,4096 -npl 1,2,4 -ntg 32"
ARGS_B="-d 0 -p 2048 -n 32"
fi
if (( DIO )); then
ARGS_BB="${ARGS_BB} --no-mmap --direct-io"
ARGS_B="${ARGS_B} -mmp 0 -dio 1"
fi
run_model() {
local HFR=$1
local HFF=$2
local HFR=$1
local HFF=$2
printf "## ${HFR}\n" | tee -a "$RESULTS"
printf "\n" | tee -a "$RESULTS"
printf "Model: https://huggingface.co/${HFR}\n" | tee -a "$RESULTS"
printf "\n" | tee -a "$RESULTS"
printf "## ${HFR}\n" | tee -a "$RESULTS"
printf "\n" | tee -a "$RESULTS"
printf "Model: https://huggingface.co/${HFR}\n" | tee -a "$RESULTS"
printf "\n" | tee -a "$RESULTS"
printf -- "- \`llama-batched-bench\`\n" | tee -a "$RESULTS"
printf "\n" | tee -a "$RESULTS"
printf -- "- \`llama-batched-bench\`\n" | tee -a "$RESULTS"
printf "\n" | tee -a "$RESULTS"
./bin/llama-batched-bench \
-hfr "${HFR}" -hff "${HFF}" \
-m "${HFF}" -fa 1 -ub 2048 --no-mmap \
${ARGS_BB} | tee -a "$RESULTS"
./bin/llama-batched-bench \
-hfr "${HFR}" -hff "${HFF}" \
-m "${HFF}" -fa 1 -ub 2048 \
${ARGS_BB} | tee -a "$RESULTS"
printf "\n" | tee -a "$RESULTS"
printf "\n" | tee -a "$RESULTS"
printf -- "- \`llama-bench\`\n" | tee -a "$RESULTS"
printf "\n" | tee -a "$RESULTS"
printf -- "- \`llama-bench\`\n" | tee -a "$RESULTS"
printf "\n" | tee -a "$RESULTS"
./bin/llama-bench \
-m "${HFF}" -fa 1 -ub 2048 -mmp 0 \
${ARGS_B} | tee -a "$RESULTS"
./bin/llama-bench \
-m "${HFF}" -fa 1 -ub 2048 \
${ARGS_B} | tee -a "$RESULTS"
printf "\n" | tee -a "$RESULTS"
printf "\n" | tee -a "$RESULTS"
printf "\n"
printf "\n"
}
run_model "ggml-org/gpt-oss-20b-GGUF" "gpt-oss-20b-mxfp4.gguf"
@ -55,6 +62,7 @@ run_model "ggml-org/gpt-oss-120b-GGUF" "gpt-oss-120b-mxfp4-
run_model "ggml-org/Qwen3-Coder-30B-A3B-Instruct-Q8_0-GGUF" "qwen3-coder-30b-a3b-instruct-q8_0.gguf"
run_model "ggml-org/Qwen2.5-Coder-7B-Q8_0-GGUF" "qwen2.5-coder-7b-q8_0.gguf"
run_model "ggml-org/gemma-3-4b-it-qat-GGUF" "gemma-3-4b-it-qat-Q4_0.gguf"
run_model "ggml-org/GLM-4.7-Flash-GGUF" "GLM-4.7-Flash-Q8_0.gguf"
if [[ -f models-extra.txt ]]; then
while read -r HFR HFF; do