Open pily1 opened 2 months ago
when use this command ./llama-minicpmv-cli -m ./models/ggml-model-Q5_K_M.gguf --mmproj ./models/mmproj-model-f16.gguf -c 4096 --temp 0.7 --top-p 0.8 --top-k 100 --repeat-penalty 1.05 --video ./test.mp4 -p "please descripe the video"
it outputs the error error:unknown argument: --video
output the description of the video
./llama-minicpmv-cli -m ./models/ggml-model-Q5_K_M.gguf --mmproj ./models/mmproj-model-f16.gguf -c 4096 --temp 0.7 --top-p 0.8 --top-k 100 --repeat-penalty 1.05 --video ./test.mp4 -p "please descripe the video"
- OS: - Python: - Transformers: - PyTorch: - CUDA (`python -c 'import torch; print(torch.version.cuda)'`):
I find the "--video" in the llama.cpp folder,I can't find it
hello,Do you use our fork of llamacpp?
是否已有关于该错误的issue或讨论? | Is there an existing issue / discussion for this?
该问题是否在FAQ中有解答? | Is there an existing answer for this in FAQ?
当前行为 | Current Behavior
when use this command ./llama-minicpmv-cli -m ./models/ggml-model-Q5_K_M.gguf --mmproj ./models/mmproj-model-f16.gguf -c 4096 --temp 0.7 --top-p 0.8 --top-k 100 --repeat-penalty 1.05 --video ./test.mp4 -p "please descripe the video"
it outputs the error error:unknown argument: --video
期望行为 | Expected Behavior
output the description of the video
复现方法 | Steps To Reproduce
./llama-minicpmv-cli -m ./models/ggml-model-Q5_K_M.gguf --mmproj ./models/mmproj-model-f16.gguf -c 4096 --temp 0.7 --top-p 0.8 --top-k 100 --repeat-penalty 1.05 --video ./test.mp4 -p "please descripe the video"
运行环境 | Environment
备注 | Anything else?
I find the "--video" in the llama.cpp folder,I can't find it