SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2408.08054 · 2024

Text2BIM: Generating Building Models Using a Large Language Model-based Multi-Agent Framework

arXiv · PDF · Open in the Atlas

Code that ran

4 of 7 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; 3 no recorded run), measured 2026-11-21; 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/2408.08054.json · badge: /api/ran/2408.08054.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
dcy0577/Text2BIM canonical (classifier label)
classifier basis: census
4 ran · 3 no recorded run
FunctionStatusWhere it lives
extract_issue_descriptions Ran
ran; our contract draft was wrong, not the code
dcy0577/Text2BIM/tool_agent/solibri_checker.py
code served (permissive licence) · get_code("4727fbb3a2745b8b")
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 dict. 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 re
from collections import Counter

# Test case 1: Simple Solibri report text with multiple issues
text_simple = """
Issue description: Wall is not properly aligned
Some other content here
Issue description: Door opening size is incorrect
More content
Issue description: Wall is not properly aligned
End of report
"""

# Test case 2: Text with edge cases
text_edge = """
Issue description: Component missing
Issue description: Component missing
Issue description: Dimension error
"""

# Test case 3: Empty/None cases
text_empty = ""

# Test case 4: Complex text with multiple occurrences
text_complex = """
Analysis Report
Issue description: Structural clash detected
Issue description: Material specification missing
Issue description: Structural clash detected
Issue description: Dimension out of tolerance
Issue description: Material specification missing
Issue description: Material specification missing
Footer information
"""

ARGS = {
    'text': text_complex
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
local_image_to_data_url Ran
ran; our contract draft was wrong, not the code
dcy0577/Text2BIM/tool_agent/utils.py
pointer only (licence: MIT); the stored copy get_code serves, from zwq2018/multi-modal-self-instruct, is pointer only (licence: NONE) · get_code("dd830bd5206c5822")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from zwq2018/multi-modal-self-instruct, and GitHub detected no licence in that repository when it was checked. 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.

remove_last_human_message_with_regex Ran
ran; our contract draft was wrong, not the code
dcy0577/Text2BIM/tool_agent/utils.py
code served (permissive licence) · get_code("48439bc74751aec6")
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 str. 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 re

def remove_last_human_message_with_regex(text):
    pattern = '\\s*\\nUser: [^\\n]*\\s*\\Z|\\A\\s*User: [^\\n]*\\s*\\Z'
    new_text = re.sub(pattern, '', text, flags=re.DOTALL).strip()
    return new_text

# Test case 1: text with User message at the end
text_1 = "Assistant: Hello\nUser: How are you?"

# Test case 2: text with User message at the beginning
text_2 = "User: Hello there\nAssistant: Hi!"

# Test case 3: text with User message at the end with whitespace
text_3 = "Assistant: Response\nUser: Question  \n"

# Test case 4: text without User message
text_4 = "Assistant: Just a response"

# Test case 5: text with multiple messages, User at end
text_5 = "User: First\nAssistant: Response\nUser: Second?"

ARGS = {
    "text": text_1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
safe_encode Ran
ran; our contract draft was wrong, not the code
dcy0577/Text2BIM/tool_agent/utils.py
code served (permissive licence) · get_code("bf6d6090ecb2723d")
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 dict. 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 types

torch.manual_seed(0)

# Test various object types that safe_encode might receive

# 1. A module type (should return None)
module_obj = types.ModuleType("test_module")

# 2. A class with __dict__ attribute
class CustomClass:
    def __init__(self):
        self.value = 42
        self.name = "test"

custom_instance = CustomClass()

# 3. A tensor (will fall back to str() since tensors don't have __dict__)
tensor_obj = torch.randn(2, 4)

# 4. A simple string (will use str())
string_obj = "hello"

# 5. An integer (will use str())
int_obj = 123

# 6. A dict (has __dict__ attribute)
dict_obj = {"key": "value"}

# 7. A list (will use str() since it doesn't have __dict__)
list_obj = [1, 2, 3]

# 8. A torch.nn.Module instance (has __dict__)
class SimpleModel(torch.nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = torch.nn.Linear(4, 2)
    
    def forward(self, x):
        return self.linear(x)

model_obj = SimpleModel()

# Create ARGS dict with one representative object to test the function
ARGS = {
    "obj": custom_instance
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
evaluate Not yet run
unverified (harvested, no recorded run)
dcy0577/Text2BIM/tool_agent/python_interpreter.py
code served (permissive licence) · get_code("623800324f32e3a0")
How it ran

no invocation recorded.

evaluate_ast Not yet run
unverified (harvested, no recorded run)
dcy0577/Text2BIM/tool_agent/python_interpreter.py
code served (permissive licence) · get_code("2738baf47117703a")
How it ran

no invocation recorded.

evaluate_while Not yet run
unverified (harvested, no recorded run)
dcy0577/Text2BIM/tool_agent/python_interpreter.py
code served (permissive licence) · get_code("be79bb275d866d66")
How it ran

no invocation recorded.

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 1 later) and 6 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
local_image_to_data_url 4 other papers
3 dated earlier, 1 later
2311.16502 (2023-11)
MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI
2406.19643 (2024-06)
Debate-to-Write: A Persona-Driven Multi-Agent Framework for Diverse Argument Generation
2407.07053 (2024-07)
Multimodal Self-Instruct: Synthetic Abstract Image and Visual Reasoning Instruction Using Language Model
and 1 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 2026-11-21.

Repositories linked to this paper

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

Abstract

The conventional BIM authoring process typically requires designers to master complex and tedious modeling commands in order to materialize their design intentions within BIM authoring tools. This additional cognitive burden complicates the design process and hinders the adoption of BIM and model-based design in the AEC (Architecture, Engineering, and Construction) industry. To facilitate the expression of design intentions more intuitively, we propose Text2BIM, an LLM-based multi-agent framework that can generate 3D building models from natural language instructions. This framework orchestrates multiple LLM agents to collaborate and reason, transforming textual user input into imperative code that invokes the BIM authoring tool's APIs, thereby generating editable BIM models with internal layouts, external envelopes, and semantic information directly in the software. Furthermore, a rule-based model checker is introduced into the agentic workflow, utilizing predefined domain knowledge to guide the LLM agents in resolving issues within the generated models and iteratively improving model quality. Extensive experiments were conducted to compare and analyze the performance of three different LLMs under the proposed framework. The evaluation results demonstrate that our approach can effectively generate high-quality, structurally rational building models that are aligned with the abstract concepts specified by user input. Finally, an interactive software prototype was developed to integrate the framework into the BIM authoring software Vectorworks, showcasing the potential of modeling by chatting. The code is available at: https://github.com/dcy0577/Text2BIM

For agents

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

get_harvested_code_for_paper("2408.08054")
get_code_for_paper("2408.08054")
have("2408.08054")

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

curl https://syntology.ai/api/ran/2408.08054.json

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

[![Syntology run record](https://syntology.ai/api/ran/2408.08054.svg)](https://syntology.ai/paper/2408.08054)

Connect an agent — have() is free.