Failed to allocate about 882.11 MiB vram on a video card which has more than 2G of free vram

[playgithub]

PROBLEM

It failed to allocate about 882.11 MiB vram on a video card which has more than 2G of free vram.
What's the problem?

BTW it works for the medium model "ggml-medium-q5_0.bin".

LOGS

$ /path/to/whisper.cpp/build-gpu/bin/main -m ~/Downloads/ggml-large-v3-q5_0.bin ~/Downloads/untitled.wav
...
ggml_cuda_init: found 1 CUDA devices:
  Device 0: NVIDIA GeForce GTX 1650, compute capability 7.5, VMM: yes
whisper_model_load:    CUDA0 total size =  1080.58 MB
whisper_model_load: model size    = 1080.47 MB
whisper_backend_init_gpu: using CUDA backend
whisper_init_state: kv self size  =  251.66 MB
whisper_init_state: kv cross size =  251.66 MB
whisper_init_state: kv pad  size  =    7.86 MB
whisper_init_state: compute buffer (conv)   =   36.13 MB
ggml_backend_cuda_buffer_type_alloc_buffer: allocating 882.11 MiB on device 0: cudaMalloc failed: out of memory
ggml_gallocr_reserve_n: failed to allocate CUDA0 buffer of size 924960000
Segmentation fault (core dumped)

$ glxinfo -B
name of display: :0
display: :0  screen: 0
direct rendering: Yes
Memory info (GL_NVX_gpu_memory_info):
    Dedicated video memory: 4096 MB
    Total available memory: 4096 MB
    Currently available dedicated video memory: 2050 MB
OpenGL vendor string: NVIDIA Corporation
OpenGL renderer string: NVIDIA GeForce GTX 1650/PCIe/SSE2
OpenGL core profile version string: 4.6.0 NVIDIA 555.58.02
OpenGL core profile shading language version string: 4.60 NVIDIA
OpenGL core profile context flags: (none)
OpenGL core profile profile mask: core profile

OpenGL version string: 4.6.0 NVIDIA 555.58.02
OpenGL shading language version string: 4.60 NVIDIA
OpenGL context flags: (none)
OpenGL profile mask: (none)

OpenGL ES profile version string: OpenGL ES 3.2 NVIDIA 555.58.02
OpenGL ES profile shading language version string: OpenGL ES GLSL ES 3.20