SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2312.07488 · 2023

LMDrive: Closed-Loop End-to-End Driving with Large Language Models

arXiv · PDF · Open in the Atlas

Code that ran

7 of 8 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 2 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 5 failures of our instrument; 1 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/2312.07488.json · badge: /api/ran/2312.07488.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
opendilab/lmdrive canonical (classifier label)
classifier basis: census
7 ran · 1 no recorded run
FunctionStatusWhere it lives
add_rect Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
opendilab/lmdrive/vision_encoder/render.py
code served (permissive licence) · get_code("a8458935002f9648")
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 ndarray, ndim 3, dtype family int, shape (100, 100, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape 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 numpy as np
import cv2

# Create a small test image (100x100, 3 channels for BGR)
img = np.ones((100, 100, 3), dtype=np.uint8) * 255

# Location: 2D point (x, y) in world coordinates
loc = np.array([0.0, 0.0])

# Orientation: unit vector (normalized)
ori = np.array([1.0, 0.0])

# Box dimensions: [horizontal_extent, vertical_extent]
box = np.array([0.5, 0.3])

# Value: scalar multiplier for color intensity
value = 0.8

# Pixels per meter: scale factor
pixels_per_meter = 10

# Max distance: half-width of the world in meters
max_distance = 5.0

# Color: RGB color values (will be scaled by value)
color = np.array([255, 0, 0], dtype=np.uint8)

ARGS = {
    "img": img,
    "loc": loc,
    "ori": ori,
    "box": box,
    "value": value,
    "pixels_per_meter": pixels_per_meter,
    "max_distance": max_distance,
    "color": color,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family int, shape (100, 100, 3) — shape (100, 100, 3) as a later run recorded it
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
convert_grid_to_xy Ran
ran, honoured the contract we drafted
opendilab/lmdrive/vision_encoder/render.py
code served (permissive licence) · get_code("ec66d5c2bef0fece")
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 float; 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)

# The function convert_grid_to_xy takes two parameters: i and j
# Both are used in simple arithmetic operations (subtraction)
# They appear to be numeric values representing grid indices

i = 5
j = 30

ARGS = {
    "i": i,
    "j": j,
}
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 float — 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
find_peak_box Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
opendilab/lmdrive/vision_encoder/render.py
code served (permissive licence) · get_code("69493ac7629f05fe")
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 list, ndim 2, 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 numpy as np

# The function expects a 3D numpy array with shape (50, 50, 8)
# It will be embedded into a (52, 52, 8) array with padding
# The first channel (index 0) contains detection confidence scores
# that are compared against thresholds (0.9 and 0.4) and neighbors

data = np.random.rand(50, 50, 8).astype(np.float32)

# Create some peaks by setting high values in the first channel
# at certain locations so the function can find them
data[5, 5, 0] = 0.95  # Peak with value > 0.9
data[10, 10, 0] = 0.5
data[10, 9, 0] = 0.3
data[10, 11, 0] = 0.3
data[9, 10, 0] = 0.3
data[11, 10, 0] = 0.3  # Local maximum with value > 0.4

data[20, 20, 0] = 0.92  # Another peak > 0.9

ARGS = {
    "data": data
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 2, dtype family int — a list 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
generate_caption Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
opendilab/lmdrive/LAVIS/app/caption.py
code served (permissive licence) · get_code("56f02812d66a0d11")
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 list, ndim 1, dtype family float; 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
import torch.nn as nn

torch.manual_seed(0)

class MockCaptionModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 8)
    
    def generate(self, samples, use_nucleus_sampling=False, num_beams=3, max_length=40, min_length=5, top_p=0.9):
        # Mock generate method that returns a list of captions
        # Each caption is a string
        batch_size = samples['image'].shape[0]
        captions = ["a cat on a mat"] * batch_size
        return captions

# Create a small image tensor (batch_size=2, channels=3, height=4, width=4)
image = torch.randn(2, 3, 4, 4)

# Create the model
model = MockCaptionModel()

ARGS = {
    'model': model,
    'image': image,
    'use_nucleus_sampling': False,
    'num_beams': 3,
    'max_length': 40,
    'min_length': 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 list, ndim 1, dtype family float — a list 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
get_actor_display_name Ran
ran; our contract draft was wrong, not the code
opendilab/lmdrive/scenario_runner/manual_control.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from findaheng/Scenic, is pointer only (licence: NOASSERTION) · get_code("c882c504d11f2b36")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from findaheng/Scenic, and GitHub could not classify the licence of that repository (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

read_img Ran
ran, honoured the contract we drafted
opendilab/lmdrive/LAVIS/app/calculate_coco_features.py
code served (permissive licence) · get_code("13a43a7d815937e6")
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 Image. 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 tempfile
import os
from PIL import Image

# Create a temporary image file for testing
temp_dir = tempfile.mkdtemp()
temp_image_path = os.path.join(temp_dir, "test_image.png")

# Create a simple RGB image and save it
test_image = Image.new('RGB', (4, 3), color=(255, 0, 0))
test_image.save(temp_image_path)

ARGS = {
    'filepath': temp_image_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Image — a Image output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Image output the recorded ndim 3 and dtype family int 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
retransform Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
opendilab/lmdrive/vision_encoder/train_pretrain.py
code served (permissive licence) · get_code("42b372b019c0d5a0")
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 Tensor, ndim 4, dtype family float, shape (2, 3, 4, 5) as a later run recorded it. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type, ndim and shape 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)

# The function expects a tensor that has been normalized
# It will denormalize it by multiplying by std and adding mean
# The tensor should be on CPU (we'll let the function fail on cuda() if needed)
# Shape should be (batch, channels, height, width) or similar with 3 channels
# Using shape (2, 3, 4, 5) - batch=2, channels=3, height=4, width=5

data = torch.randn(2, 3, 4, 5)

ARGS = {
    "data": data
}
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 Tensor, ndim 4, dtype family float, shape (2, 3, 4, 5) — shape (2, 3, 4, 5) as a later run recorded it
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
generate_script Not yet run
unverified (harvested, no recorded run)
opendilab/lmdrive/data_collection/generate_bashs.py
code served (permissive licence) · get_code("413cd3fef973bb3d")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 2 the same month, 3 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
generate_caption 12 other papers
4 dated earlier, 2 the same month, 6 later
2305.01278 (2023-05)
VPGTrans: Transfer Visual Prompt Generator across LLMs
2305.06988 (2023-05)
Self-Chained Image-Language Model for Video Localization and Question Answering
2307.07063 (2023-07)
Bootstrapping Vision-Language Learning with Decoupled Language Pre-training
and 9 more not named here
read_img 12 other papers
4 dated earlier, 2 the same month, 6 later
2305.01278 (2023-05)
VPGTrans: Transfer Visual Prompt Generator across LLMs
2305.06988 (2023-05)
Self-Chained Image-Language Model for Video Localization and Question Answering
2307.07063 (2023-07)
Bootstrapping Vision-Language Learning with Decoupled Language Pre-training
and 9 more not named here
get_actor_display_name 8 other papers
5 dated earlier, 3 later
1904.08980 (2019-04)
Exploring the Limitations of Behavior Cloning for Autonomous Driving
1911.04175 (2019-11)
Multi-Agent Connected Autonomous Driving using Deep Reinforcement Learning
1911.10868 (2019-11)
End-to-End Model-Free Reinforcement Learning for Urban Driving using Implicit Affordances
and 5 more not named here

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

Despite significant recent progress in the field of autonomous driving, modern methods still struggle and can incur serious accidents when encountering long-tail unforeseen events and challenging urban scenarios. On the one hand, large language models (LLM) have shown impressive reasoning capabilities that approach "Artificial General Intelligence". On the other hand, previous autonomous driving methods tend to rely on limited-format inputs (e.g. sensor data and navigation waypoints), restricting the vehicle's ability to understand language information and interact with humans. To this end, this paper introduces LMDrive, a novel language-guided, end-to-end, closed-loop autonomous driving framework. LMDrive uniquely processes and integrates multi-modal sensor data with natural language instructions, enabling interaction with humans and navigation software in realistic instructional settings. To facilitate further research in language-based closed-loop autonomous driving, we also publicly release the corresponding dataset which includes approximately 64K instruction-following data clips, and the LangAuto benchmark that tests the system's ability to handle complex instructions and challenging driving scenarios. Extensive closed-loop experiments are conducted to demonstrate LMDrive's effectiveness. To the best of our knowledge, we're the very first work to leverage LLMs for closed-loop end-to-end autonomous driving. Codes, models, and datasets can be found at https://github.com/opendilab/LMDrive

For agents

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

get_harvested_code_for_paper("2312.07488")
get_code_for_paper("2312.07488")
have("2312.07488")

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

curl https://syntology.ai/api/ran/2312.07488.json

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

[![Syntology run record](https://syntology.ai/api/ran/2312.07488.svg)](https://syntology.ai/paper/2312.07488)

Connect an agent — have() is free.