SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2410.18558 · 2024

Infinity-MM: Scaling Multimodal Performance with Large-Scale and High-Quality Instruction Data

arXiv · PDF · Open in the Atlas

Code that ran

5 of 6 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 5 have no contract verdict; 0 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/2410.18558.json · badge: /api/ran/2410.18558.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
flagopen/flagscale community (classifier label)
classifier basis not recorded on the link
5 ran · 1 no recorded run
FunctionStatusWhere it lives
flatten_dict_to_args Ran
ran on a synthesized input
flagopen/flagscale/flagscale/utils.py
code served (permissive licence) · get_code("a946897a635a9ffa")
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 numpy as np

torch.manual_seed(0)

ARGS = {
    'config_dict': {
        'model_name': 'bert-base',
        'learning_rate': 0.001,
        'batch_size': 32,
        'use_cuda': True,
        'layers': [64, 128, 256],
        'nested_config': {
            'hidden_dim': 512,
            'dropout': 0.1
        },
        'skip_this': 'ignored_value'
    },
    'ignore_keys': ['skip_this']
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, 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
make_headers Ran
ran on a synthesized input
flagopen/flagscale/flagscale/eval/robo.py
code served (permissive licence) · get_code("6c1f969a5ed98237")
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 hashlib
import hmac
import time

# Set up arguments for make_headers function
url = "https://api.example.com/endpoint"
body = '{"key": "value", "data": [1, 2, 3]}'
secret = "my_secret_key_12345"
include_content_type = True

ARGS = {
    "url": url,
    "body": body,
    "secret": secret,
    "include_content_type": include_content_type
}
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
prepare_compress_methods Ran
ran on a synthesized input
flagopen/flagscale/flagscale/compress/combined_algo.py
code served (permissive licence) · get_code("9dd40bc51eb5b8b0")
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 copy
import torch

torch.manual_seed(0)

# The function takes compress_cfg and does a deepcopy on it
# compress_cfg should be a dict-like configuration object
# Based on the function body, it just needs to be deepcopyable

compress_cfg = {
    "quantization": {
        "bits": 8,
        "method": "symmetric"
    },
    "pruning": {
        "sparsity": 0.5,
        "layer_types": ["Linear", "Conv2d"]
    },
    "distillation": {
        "temperature": 4.0,
        "alpha": 0.5
    }
}

ARGS = {
    "compress_cfg": compress_cfg
}
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
prepare_config Ran
ran on a synthesized input
flagopen/flagscale/flagscale/compress/compressor.py
code served (permissive licence) · get_code("ac0e60d70c14acfa")
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 DictConfig. 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
import yaml
from omegaconf import OmegaConf

# Create a temporary YAML config file with the expected structure
temp_dir = tempfile.mkdtemp()
config_path = os.path.join(temp_dir, "test_config.yaml")

config_content = {
    'experiment': 'test_experiment',
    'action': 'train',
    'model': {
        'name': 'test_model',
        'hidden_dim': 128,
        'num_layers': 2
    },
    'training': {
        'learning_rate': 0.001,
        'batch_size': 32,
        'epochs': 10
    },
    'data': {
        'dataset_name': 'cifar10',
        'num_workers': 4
    }
}

with open(config_path, 'w') as f:
    yaml.dump(config_content, f)

ARGS = {
    'config_path': config_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
omegaconf, yaml — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DictConfig — a DictConfig output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DictConfig 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
start_server Ran
ran on a synthesized input
flagopen/flagscale/flagscale/eval/robo.py
code served (permissive licence) · get_code("e580a09ddd5d3d19")
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 Popen. 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 subprocess
import sys

# The function uses model_name as a string parameter
# It formats it into a path and passes it to a subprocess call
# We just need a valid model name string

ARGS = {
    'model_name': 'test_model'
}
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 Popen — a Popen output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Popen 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
auto_remote Not yet run
unverified (harvested, no recorded run)
flagopen/flagscale/flagscale/serve/core.py
code served (permissive licence) · get_code("0d33aaf712b41133")
How it ran

no invocation recorded.

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 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 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.

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

Recently, Vision-Language Models (VLMs) have achieved remarkable progress in multimodal tasks, and multimodal instruction data serves as the foundation for enhancing VLM capabilities. Despite the availability of several open-source multimodal datasets, limitations in the scale and quality of open-source instruction data hinder the performance of VLMs trained on these datasets, leading to a significant gap compared to models trained on closed-source data. To address this challenge, we introduce Infinity-MM, a large-scale multimodal instruction dataset. We collected the available multimodal instruction datasets and performed unified preprocessing, resulting in a dataset with over 40 million samples that ensures diversity and accuracy. Furthermore, to enable large-scale expansion of instruction data and support the continuous acquisition of high-quality data, we propose a synthetic instruction generation method based on a tagging system and open-source VLMs. By establishing correspondences between different types of images and associated instruction types, this method can provide essential guidance during data synthesis. Leveraging this high-quality data, we have trained a 2-billion-parameter Vision-Language Model, Aquila-VL-2B, which achieves state-of-the-art (SOTA) performance among models of similar scale. The data is available at: https://huggingface.co/datasets/BAAI/Infinity-MM.

For agents

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

get_harvested_code_for_paper("2410.18558")
get_code_for_paper("2410.18558")
have("2410.18558")

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

curl https://syntology.ai/api/ran/2410.18558.json

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

[![Syntology run record](https://syntology.ai/api/ran/2410.18558.svg)](https://syntology.ai/paper/2410.18558)

Connect an agent — have() is free.