SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2408.01584 · ICLR · 2025

GPUDrive: Data-driven, multi-agent driving simulation at 1 million FPS

Eugene Vinitsky, Brennan Shacklett, Saman Kazemkhani, Aarav Pandya, Daphne Cornelisse

arXiv · PDF · Open in the Atlas

Code that ran

4 of 4 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; 0 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 4 failures of our instrument; 0 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/2408.01584.json · badge: /api/ran/2408.01584.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
Emerge-Lab/gpudrive canonical (classifier label)
classifier basis: census
4 ran · 0 no recorded run
FunctionStatusWhere it lives
entropy Ran
ran; our contract draft was wrong, not the code
Emerge-Lab/gpudrive/gpudrive/networks/late_fusion.py
code served (permissive licence) · get_code("140d1fb7fc50b991")
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 1, dtype family float, shape (2,) 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 torch
from torch.distributions.utils import logits_to_probs

torch.manual_seed(0)

# The function entropy takes logits as input
# logits should be a tensor that can be clamped and used in arithmetic operations
# The function applies logits_to_probs which expects a tensor
# and performs operations along the last dimension (-1)

# Create a small tensor with shape (2, 4) - batch size 2, dimension 4
logits = torch.randn(2, 4)

ARGS = {
    "logits": logits
}
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 1, dtype family float, shape (2,) — shape (2,) 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
log_prob Ran
ran; our contract draft was wrong, not the code
Emerge-Lab/gpudrive/gpudrive/networks/late_fusion.py
code served (permissive licence) · get_code("54928c52a27bfaa2")
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 1, dtype family float, shape (2,) 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 torch

torch.manual_seed(0)

# Construct arguments for log_prob function
# logits: tensor of shape (..., num_classes) representing log probabilities
# value: tensor of shape (...,) containing class indices

logits = torch.randn(2, 4)  # batch_size=2, num_classes=4
value = torch.tensor([1, 3])  # batch_size=2, class indices

ARGS = {
    "logits": logits,
    "value": value,
}
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 1, dtype family float, shape (2,) — shape (2,) 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
merge_actions Ran
ran; our contract draft was wrong, not the code
Emerge-Lab/gpudrive/gpudrive/agents/core.py
code served (permissive licence) · get_code("268080dd30ffe9fe")
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 2, dtype family int, shape (2, 3) 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)

# Construct actor_actions_dict
# Maps actor names to tensors of actions, one per world
actor_actions_dict = {
    'actor_1': torch.tensor([[0], [1]]),  # shape (2 worlds, 1 action per world)
    'actor_2': torch.tensor([[2], [3]]),  # shape (2 worlds, 1 action per world)
}

# Construct actor_ids_dict
# Maps actor names to lists of indices per world
actor_ids_dict = {
    'actor_1': [[0], [1]],  # actor_1 controls agent 0 in world 0, agent 1 in world 1
    'actor_2': [[1], [0]],  # actor_2 controls agent 1 in world 0, agent 0 in world 1
}

# Construct reference_action_tensor
# Shape: (num_worlds=2, max_num_controllable_agents=3)
reference_action_tensor = torch.zeros((2, 3), dtype=torch.long)

ARGS = {
    'actor_actions_dict': actor_actions_dict,
    'actor_ids_dict': actor_ids_dict,
    'reference_action_tensor': reference_action_tensor,
    'verbose': False,
    'device': 'cpu',
}
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 2, dtype family int, shape (2, 3) — shape (2, 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
sample_logits Ran
ran; our contract draft was wrong, not the code
Emerge-Lab/gpudrive/gpudrive/networks/late_fusion.py
code served (permissive licence) · get_code("c5743ad5afe917b0")
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 2, 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
from torch.distributions.utils import logits_to_probs

torch.manual_seed(0)

# Create small logits tensor: batch_size=2, num_actions=4
logits_tensor = torch.randn(2, 4)

# Test with single tensor (no action, deterministic=False)
ARGS = {
    "logits": logits_tensor,
    "action": None,
    "deterministic": 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 tuple, ndim 2, 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

Also present in other papers' code

Of 4 samples lifted from this paper's linked repositories, 4 carry a function identity and 0 do not. Of the 4 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 4 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.

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

Multi-agent learning algorithms have been successful at generating superhuman planning in various games but have had limited impact on the design of deployed multi-agent planners. A key bottleneck in applying these techniques to multiagent planning is that they require billions of steps of experience. To enable the study of multi-agent planning at scale, we present GPUDrive. GPUDrive is a GPU-accelerated, multi-agent simulator built on top of the Madrona Game Engine capable of generating over a million simulation steps per second. Observation, reward, and dynamics functions are written directly in C++, allowing users to define complex, heterogeneous agent behaviors that are lowered to highperformance CUDA. Despite these low-level optimizations, GPUDrive is fully accessible through Python, offering a seamless and efficient workflow for multiagent, closed-loop simulation. Using GPUDrive, we train reinforcement learning agents on the Waymo Open Motion Dataset, achieving efficient goal-reaching in minutes and scaling to thousands of scenarios in hours. We open-source the code and pre-trained agents at www.github.com/Emerge-Lab/gpudrive

For agents

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

get_harvested_code_for_paper("2408.01584")
get_code_for_paper("2408.01584")
have("2408.01584")

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

curl https://syntology.ai/api/ran/2408.01584.json

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

[![Syntology run record](https://syntology.ai/api/ran/2408.01584.svg)](https://syntology.ai/paper/2408.01584)

Connect an agent — have() is free.