SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2403.06977 · ECCV · 2024

VideoMamba: State Space Model for Efficient Video Understanding

Yu Qiao, Kunchang Li, Yali Wang, Limin Wang, Xinhao Li

arXiv · PDF · Open in the Atlas

Code that ran

9 of 15 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (5 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 0 violated it and 5 have no contract verdict; 3 failures of our instrument; 6 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2403.06977.json · badge: /api/ran/2403.06977.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
hongyuanyu/span community
Papers with Code archive
6 ran · 2 no recorded run
opengvlab/videomamba marked official in Papers with Code
Papers with Code archive
3 ran · 4 no recorded run
FunctionStatusWhere it lives
Block Ran
ran on a synthesized input · constructed an object; no result was computed
opengvlab/videomamba/videomamba/video_sm/models/videomamba.py
code served (permissive licence) · get_code("f656db702b49f748")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, Block: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
from torch import Tensor
from typing import Optional

torch.manual_seed(0)

# Define a minimal mixer class that satisfies the interface
class SimpleMixer(nn.Module):
    def __init__(self, dim):
        super().__init__()
        self.dim = dim
        self.linear = nn.Linear(dim, dim)
    
    def forward(self, hidden_states: Tensor, inference_params=None):
        return self.linear(hidden_states)
    
    def allocate_inference_cache(self, batch_size, max_seqlen, dtype=None, **kwargs):
        return None

# Create a small tensor for testing
hidden_states = torch.randn(2, 3, 4)  # batch=2, seq_len=3, dim=4

# Define all arguments
ARGS = {
    'dim': 4,
    'mixer_cls': SimpleMixer,
    'norm_cls': nn.LayerNorm,
    'fused_add_norm': False,
    'residual_in_fp32': False,
    'drop_path': 0.0,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
timm, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Block — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a Block output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
Conv3XC Ran
ran on a synthesized input · constructed an object; no result was computed
hongyuanyu/span/basicsr/archs/span_arch.py
code served (permissive licence) · get_code("d1973755b9d2f056")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, Conv3XC: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# Conv3XC is a nn.Module that performs convolution operations
# Parameters:
# - c_in: input channels
# - c_out: output channels
# - gain1: gain factor (default 1)
# - gain2: gain factor (default 0, appears unused in code)
# - s: stride (default 1)
# - bias: whether to use bias (default True)
# - relu: whether to use relu (default False)

ARGS = {
    'c_in': 3,
    'c_out': 4,
    'gain1': 1,
    'gain2': 0,
    's': 1,
    'bias': True,
    'relu': False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Conv3XC — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a Conv3XC output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
Mamba Ran
ran on a synthesized input · constructed an object; no result was computed
opengvlab/videomamba/videomamba/video_sm/models/videomamba.py
code served (permissive licence) · get_code("d83cab7852a90b13")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, Mamba: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
import math

torch.manual_seed(0)

ARGS = {
    'd_model': 4,
    'd_state': 16,
    'd_conv': 4,
    'expand': 2,
    'dt_rank': 'auto',
    'dt_min': 0.001,
    'dt_max': 0.1,
    'dt_init': 'random',
    'dt_scale': 1.0,
    'dt_init_floor': 0.0001,
    'conv_bias': True,
    'bias': False,
    'use_fast_path': True,
    'layer_idx': 0,
    'device': None,
    'dtype': None,
    'bimamba': True,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
einops, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Mamba — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a Mamba output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
PatchEmbed Ran
ran on a synthesized input · constructed an object; no result was computed
opengvlab/videomamba/videomamba/video_sm/models/videomamba.py
code served (permissive licence) · get_code("f9b7873ac04f0368")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, PatchEmbed: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
from timm.models.layers import DropPath, to_2tuple

torch.manual_seed(0)

ARGS = {
    "img_size": 224,
    "patch_size": 16,
    "kernel_size": 1,
    "in_chans": 3,
    "embed_dim": 768,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
timm, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type PatchEmbed — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a PatchEmbed output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
SPAB Ran
ran on a synthesized input · constructed an object; no result was computed
hongyuanyu/span/basicsr/archs/span_arch.py
code served (permissive licence) · get_code("427fdb7d52e6e55f")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, SPAB: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# SPAB expects:
# - in_channels: int
# - mid_channels: int or None
# - out_channels: int or None
# - bias: bool

ARGS = {
    'in_channels': 4,
    'mid_channels': 8,
    'out_channels': 4,
    'bias': False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type SPAB — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a SPAB output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
_make_pair Ran
ran, honoured the contract we drafted
hongyuanyu/span/basicsr/archs/span_arch.py
code served (permissive licence) · get_code("dfda97b5629cad1d")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 1, dtype family int; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

ARGS = {
    "value": 5
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family int — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
activation Ran
ran; our contract draft was wrong, not the code
hongyuanyu/span/basicsr/archs/span_arch.py
code served (permissive licence) · get_code("a2eac1cf4f3c5137")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ReLU. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
from torch import nn

torch.manual_seed(0)

ARGS = {
    'act_type': 'relu',
    'inplace': True,
    'neg_slope': 0.05,
    'n_prelu': 1,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ReLU — a ReLU output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ReLU output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
conv_layer Ran
ran; our contract draft was wrong, not the code
hongyuanyu/span/basicsr/archs/span_arch.py
code served (permissive licence) · get_code("8faca1aa2666657d")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Conv2d. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    "in_channels": 3,
    "out_channels": 16,
    "kernel_size": 3,
    "bias": True,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Conv2d — a Conv2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Conv2d output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
pixelshuffle_block Ran
ran; our contract draft was wrong, not the code
hongyuanyu/span/basicsr/archs/span_arch.py
code served (permissive licence) · get_code("3a2faf2e05a448e3")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Sequential. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
from collections import OrderedDict

torch.manual_seed(0)

ARGS = {
    'in_channels': 3,
    'out_channels': 64,
    'upscale_factor': 2,
    'kernel_size': 3,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
SPAN Not yet run
unverified (harvested, no recorded run)
hongyuanyu/span/basicsr/archs/span_arch.py
code served (permissive licence) · get_code("c87c97b8024c0fde")
How it ran

no invocation recorded.

VisionMamba Not yet run
unverified (harvested, no recorded run)
opengvlab/videomamba/videomamba/video_sm/models/videomamba.py
code served (permissive licence) · get_code("3bef289c82d678ec")
How it ran

no invocation recorded.

_init_weights Not yet run
unverified (harvested, no recorded run)
opengvlab/videomamba/videomamba/video_sm/models/videomamba.py
code served (permissive licence) · get_code("547c8f8405c4d450")
How it ran

no invocation recorded.

create_block Not yet run
unverified (harvested, no recorded run)
opengvlab/videomamba/videomamba/video_sm/models/videomamba.py
code served (permissive licence) · get_code("6a1367264702ad24")
How it ran

no invocation recorded.

segm_init_weights Not yet run
unverified (harvested, no recorded run)
opengvlab/videomamba/videomamba/video_sm/models/videomamba.py
code served (permissive licence) · get_code("d7eeadc33187dd9e")
How it ran

no invocation recorded.

sequential Not yet run
unverified (harvested, no recorded run)
hongyuanyu/span/basicsr/archs/span_arch.py
code served (permissive licence) · get_code("28179fe9fd01ec78")
How it ran

no invocation recorded.

Also present in other papers' code

Of 15 samples lifted from this paper's linked repositories, 15 carry a function identity and 0 do not. Of the 15 distinct functions with an identity, 10 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 0 the same month, 8 later) and 5 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
sequential
brick of this paper's assembly: helper
6 other papers
5 dated earlier, 1 later · 3 only as an assembly brick
2104.03078 (2021-04)
Universal and Flexible Optical Aberration Correction Using Deep-Prior Based Deconvolution
2108.05302 (2021-08)
Mutual Affine Network for Spatially Variant Kernel Estimation in Blind Image Super-Resolution
2109.14573 (2021-09)
Towards Flexible Blind JPEG Artifacts Removal
and 3 more not named here
activation
brick of this paper's assembly: helper
2 other papers
1 dated earlier, 1 later
2205.07514 (2022-05)
Residual Local Feature Network for Efficient Super-Resolution
Block
brick of this paper's assembly: model
1 other paper
1 later · 1 only as an assembly brick
2507.23595 (2025-07)
MamV2XCalib: V2X-based Target-less Infrastructure Camera Calibration with State Space Model
Mamba
brick of this paper's assembly: helper
1 other paper
1 later
2507.23595 (2025-07)
MamV2XCalib: V2X-based Target-less Infrastructure Camera Calibration with State Space Model
VisionMamba
brick of this paper's assembly: entry
1 other paper
1 later
2507.23595 (2025-07)
MamV2XCalib: V2X-based Target-less Infrastructure Camera Calibration with State Space Model
_init_weights
brick of this paper's assembly: helper
1 other paper
1 later · 1 only as an assembly brick
2507.23595 (2025-07)
MamV2XCalib: V2X-based Target-less Infrastructure Camera Calibration with State Space Model
conv_layer
brick of this paper's assembly: helper
1 other paper
1 dated earlier
2205.07514 (2022-05)
Residual Local Feature Network for Efficient Super-Resolution
create_block
brick of this paper's assembly: helper
1 other paper
1 later
2507.23595 (2025-07)
MamV2XCalib: V2X-based Target-less Infrastructure Camera Calibration with State Space Model
pixelshuffle_block
brick of this paper's assembly: helper
1 other paper
1 dated earlier
2205.07514 (2022-05)
Residual Local Feature Network for Efficient Super-Resolution
segm_init_weights
brick of this paper's assembly: helper
1 other paper
1 later · 1 only as an assembly brick
2507.23595 (2025-07)
MamV2XCalib: V2X-based Target-less Infrastructure Camera Calibration with State Space Model

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Addressing the dual challenges of local redundancy and global dependencies in video understanding, this work innovatively adapts the Mamba to the video domain. The proposed VideoMamba overcomes the limitations of existing 3D convolution neural networks (CNNs) and video transformers. Its linear-complexity operator enables efficient long-term modeling, which is crucial for high-resolution long video understanding. Extensive evaluations reveal VideoMamba's four core abilities: (1) Scalability in the visual domain without extensive dataset pretraining, thanks to a novel self-distillation technique; (2) Sensitivity for recognizing shortterm actions even with fine-grained motion differences; (3) Superiority in long-term video understanding, showcasing significant advancements over traditional feature-based models; and (4) Compatibility with other modalities, demonstrating robustness in multi-modal contexts. Through these advantages, VideoMamba sets a new benchmark, offering a scalable and efficient solution for comprehensive video understanding.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2403.06977")
get_code_for_paper("2403.06977")
have("2403.06977")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2403.06977.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2403.06977.svg)](https://syntology.ai/paper/2403.06977)

Connect an agent — have() is free.