This commit is contained in:
@@ -0,0 +1,18 @@
|
||||
# OMP_NUM_THREADS=14 please Check issue: https://github.com/AutoGPTQ/AutoGPTQ/issues/439
|
||||
OMP_NUM_THREADS=14 \
|
||||
CUDA_VISIBLE_DEVICES=0 \
|
||||
MAX_PIXELS=1003520 \
|
||||
VIDEO_MAX_PIXELS=50176 \
|
||||
FPS_MAX_FRAMES=12 \
|
||||
swift export \
|
||||
--model Qwen/Qwen2.5-Omni-7B \
|
||||
--dataset 'AI-ModelScope/alpaca-gpt4-data-zh#500' \
|
||||
'AI-ModelScope/alpaca-gpt4-data-en#500' \
|
||||
'modelscope/coco_2014_caption:validation#500' \
|
||||
'swift/VideoChatGPT:Generic#500' \
|
||||
--quant_n_samples 256 \
|
||||
--quant_batch_size 1 \
|
||||
--max_length 2048 \
|
||||
--quant_method gptq \
|
||||
--quant_bits 4 \
|
||||
--output_dir Qwen2.5-Omni-7B-GPTQ-Int4
|
||||
Reference in New Issue
Block a user