Llava-next-video got different results after using the new video processor #38221

Collecting environment information...
PyTorch version: 2.8.0.dev20250519+cpu
Is debug build: False
CUDA used to build PyTorch: None
ROCM used to build PyTorch: N/A

OS: Ubuntu 22.04.5 LTS (x86_64)
GCC version: (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
Clang version: Could not collect
CMake version: version 4.0.0
Libc version: glibc-2.35

Python version: 3.11.12 (main, Apr  9 2025, 08:55:54) [GCC 11.4.0] (64-bit runtime)
Python platform: Linux-6.11.0-21-generic-x86_64-with-glibc2.35
Is CUDA available: False
CUDA runtime version: No CUDA
CUDA_MODULE_LOADING set to: N/A
GPU models and configuration: No CUDA
Nvidia driver version: No CUDA
cuDNN version: No CUDA
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True

CPU:
Architecture:                         x86_64
CPU op-mode(s):                       32-bit, 64-bit
Address sizes:                        52 bits physical, 57 bits virtual
Byte Order:                           Little Endian
CPU(s):                               384
On-line CPU(s) list:                  0-383
Vendor ID:                            GenuineIntel
BIOS Vendor ID:                       Intel(R) Corporation
Model name:                           Intel(R) Xeon(R) 6972P
BIOS Model name:                      Intel(R) Xeon(R) 6972P
CPU family:                           6
Model:                                173
Thread(s) per core:                   2
Core(s) per socket:                   96
Socket(s):                            2
Stepping:                             1
CPU max MHz:                          3900.0000
CPU min MHz:                          800.0000
BogoMIPS:                             4800.00
Flags:                                fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi
mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nons
top_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid
dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb ca
t_l3 cat_l2 cdp_l3 intel_ppin cdp_l2 ssbd mba ibrs ibpb stibp ibrs_enhanced tpr_shadow flexpriority ept vpid ept_ad fsgsbase ts
c_adjust bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb intel_pt avx512
cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local split_lock_detect
user_shstk avx_vnni avx512_bf16 wbnoinvd dtherm ida arat pln pts hwp hwp_act_window hwp_epp hwp_pkg_req hfi vnmi avx512vbmi umi
p pku ospke waitpkg avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq la57 rdpid bus_lock_detect cld
emote movdiri movdir64b enqcmd fsrm md_clear serialize tsxldtrk pconfig arch_lbr ibt amx_bf16 avx512_fp16 amx_tile amx_int8 flu
sh_l1d arch_capabilities
Virtualization:                       VT-x
L1d cache:                            9 MiB (192 instances)
L1i cache:                            12 MiB (192 instances)
L2 cache:                             384 MiB (192 instances)
L3 cache:                             960 MiB (2 instances)
NUMA node(s):                         6
NUMA node0 CPU(s):                    0-31,192-223
NUMA node1 CPU(s):                    32-63,224-255
NUMA node2 CPU(s):                    64-95,256-287
NUMA node3 CPU(s):                    96-127,288-319
NUMA node4 CPU(s):                    128-159,320-351
NUMA node5 CPU(s):                    160-191,352-383
Vulnerability Gather data sampling:   Not affected
Vulnerability Itlb multihit:          Not affected
Vulnerability L1tf:                   Not affected
Vulnerability Mds:                    Not affected
Vulnerability Meltdown:               Not affected
Vulnerability Mmio stale data:        Not affected
Vulnerability Reg file data sampling: Not affected
Vulnerability Retbleed:               Not affected
Vulnerability Spec rstack overflow:   Not affected
Vulnerability Spec store bypass:      Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:             Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:             Mitigation; Enhanced / Automatic IBRS; IBPB conditional; RSB filling; PBRSB-eIBRS Not aff
ected; BHI BHI_DIS_S
Vulnerability Srbds:                  Not affected
Vulnerability Tsx async abort:        Not affected

Versions of relevant libraries:
[pip3] galore-torch==1.0
[pip3] numpy==1.26.4
[pip3] onnx==1.17.0
[pip3] onnxruntime==1.21.0
[pip3] optree==0.15.0
[pip3] pytorch-msssim==1.0.0
[pip3] torch==2.8.0.dev20250519+cpu
[pip3] torchao==0.11.0+git
[pip3] torchaudio==2.6.0.dev20250519+cpu
[pip3] torchvision==0.22.0.dev20250519+cpu
[pip3] triton==3.3.0
[conda] Could not collect

import av
import torch
import numpy as np
from huggingface_hub import hf_hub_download
from transformers import LlavaNextVideoProcessor, LlavaNextVideoForConditionalGeneration

model_id = "llava-hf/LLaVA-NeXT-Video-7B-hf"

model = LlavaNextVideoForConditionalGeneration.from_pretrained(
    model_id, 
    torch_dtype=torch.float16, 
    low_cpu_mem_usage=True, 
)

processor = LlavaNextVideoProcessor.from_pretrained(model_id)

def read_video_pyav(container, indices):
    '''
    Decode the video with PyAV decoder.
    Args:
        container (`av.container.input.InputContainer`): PyAV container.
        indices (`List[int]`): List of frame indices to decode.
    Returns:
        result (np.ndarray): np array of decoded frames of shape (num_frames, height, width, 3).
    '''
    frames = []
    container.seek(0)
    start_index = indices[0]
    end_index = indices[-1]
    for i, frame in enumerate(container.decode(video=0)):
        if i > end_index:
            break
        if i >= start_index and i in indices:
            frames.append(frame)
    return np.stack([x.to_ndarray(format="rgb24") for x in frames])


# define a chat history and use `apply_chat_template` to get correctly formatted prompt
# Each value in "content" has to be a list of dicts with types ("text", "image", "video") 
conversation = [
    {

        "role": "user",
        "content": [
            {"type": "text", "text": "Why is this video funny?"},
            {"type": "video"},
            ],
    },
]

prompt = processor.apply_chat_template(conversation, add_generation_prompt=True)

video_path = hf_hub_download(repo_id="raushan-testing-hf/videos-test", filename="sample_demo_1.mp4", repo_type="dataset")
container = av.open(video_path)

# sample uniformly 8 frames from the video, can sample more for longer videos
total_frames = container.streams.video[0].frames
indices = np.arange(0, total_frames, total_frames / 8).astype(int)
clip = read_video_pyav(container, indices)
inputs_video = processor(text=prompt, videos=clip, padding=True, return_tensors="pt").to(model.device)

output = model.generate(**inputs_video, max_new_tokens=100, do_sample=False)
print(processor.decode(output[0][2:], skip_special_tokens=True))

ER:
Why is this video funny? ASSISTANT: The humor in this video comes from the unexpected and endearing nature of the situation. The baby is wearing glasses and appears to be reading a book, which is a humorous and endearing sight because babies are typically not expected to be able to read at such a young age. The glasses add a touch of whimsy and the baby's expression and actions suggest that they are deeply engrossed in the book, which is a playful and amusing portray

ER:
Why is this video funny? ASSISTANT: The humor in this video comes from the unexpected and endearing situation of a young child, who appears to be a baby or toddler, attempting to read a book. The child's small size and the fact that they are reading a book that is too big for them create a comical contrast, as it's not something you would typically see a young child doing. The child's focus and determination to read the book, despite the size, adds to the humor. The